Reinforcement learning in supply chains.

PubMed

Valluri, Annapurna; North, Michael J; Macal, Charles M

2009-10-01

Effective management of supply chains creates value and can strategically position companies. In practice, human beings have been found to be both surprisingly successful and disappointingly inept at managing supply chains. The related fields of cognitive psychology and artificial intelligence have postulated a variety of potential mechanisms to explain this behavior. One of the leading candidates is reinforcement learning. This paper applies agent-based modeling to investigate the comparative behavioral consequences of three simple reinforcement learning algorithms in a multi-stage supply chain. For the first time, our findings show that the specific algorithm that is employed can have dramatic effects on the results obtained. Reinforcement learning is found to be valuable in multi-stage supply chains with several learning agents, as independent agents can learn to coordinate their behavior. However, learning in multi-stage supply chains using these postulated approaches from cognitive psychology and artificial intelligence take extremely long time periods to achieve stability which raises questions about their ability to explain behavior in real supply chains. The fact that it takes thousands of periods for agents to learn in this simple multi-agent setting provides new evidence that real world decision makers are unlikely to be using strict reinforcement learning in practice.

Seizure Control in a Computational Model Using a Reinforcement Learning Stimulation Paradigm.

PubMed

Nagaraj, Vivek; Lamperski, Andrew; Netoff, Theoden I

2017-11-01

Neuromodulation technologies such as vagus nerve stimulation and deep brain stimulation, have shown some efficacy in controlling seizures in medically intractable patients. However, inherent patient-to-patient variability of seizure disorders leads to a wide range of therapeutic efficacy. A patient specific approach to determining stimulation parameters may lead to increased therapeutic efficacy while minimizing stimulation energy and side effects. This paper presents a reinforcement learning algorithm that optimizes stimulation frequency for controlling seizures with minimum stimulation energy. We apply our method to a computational model called the epileptor. The epileptor model simulates inter-ictal and ictal local field potential data. In order to apply reinforcement learning to the Epileptor, we introduce a specialized reward function and state-space discretization. With the reward function and discretization fixed, we test the effectiveness of the temporal difference reinforcement learning algorithm (TD(0)). For periodic pulsatile stimulation, we derive a relation that describes, for any stimulation frequency, the minimal pulse amplitude required to suppress seizures. The TD(0) algorithm is able to identify parameters that control seizures quickly. Additionally, our results show that the TD(0) algorithm refines the stimulation frequency to minimize stimulation energy thereby converging to optimal parameters reliably. An advantage of the TD(0) algorithm is that it is adaptive so that the parameters necessary to control the seizures can change over time. We show that the algorithm can converge on the optimal solution in simulation with slow and fast inter-seizure intervals.

Reinforcement Learning in Distributed Domains: Beyond Team Games

NASA Technical Reports Server (NTRS)

Wolpert, David H.; Sill, Joseph; Turner, Kagan

2000-01-01

Distributed search algorithms are crucial in dealing with large optimization problems, particularly when a centralized approach is not only impractical but infeasible. Many machine learning concepts have been applied to search algorithms in order to improve their effectiveness. In this article we present an algorithm that blends Reinforcement Learning (RL) and hill climbing directly, by using the RL signal to guide the exploration step of a hill climbing algorithm. We apply this algorithm to the domain of a constellations of communication satellites where the goal is to minimize the loss of importance weighted data. We introduce the concept of 'ghost' traffic, where correctly setting this traffic induces the satellites to act to optimize the world utility. Our results indicated that the bi-utility search introduced in this paper outperforms both traditional hill climbing algorithms and distributed RL approaches such as team games.

Reinforced two-step-ahead weight adjustment technique for online training of recurrent neural networks.

PubMed

Chang, Li-Chiu; Chen, Pin-An; Chang, Fi-John

2012-08-01

A reliable forecast of future events possesses great value. The main purpose of this paper is to propose an innovative learning technique for reinforcing the accuracy of two-step-ahead (2SA) forecasts. The real-time recurrent learning (RTRL) algorithm for recurrent neural networks (RNNs) can effectively model the dynamics of complex processes and has been used successfully in one-step-ahead forecasts for various time series. A reinforced RTRL algorithm for 2SA forecasts using RNNs is proposed in this paper, and its performance is investigated by two famous benchmark time series and a streamflow during flood events in Taiwan. Results demonstrate that the proposed reinforced 2SA RTRL algorithm for RNNs can adequately forecast the benchmark (theoretical) time series, significantly improve the accuracy of flood forecasts, and effectively reduce time-lag effects.

Simultaneous vibration control and energy harvesting using actor-critic based reinforcement learning

NASA Astrophysics Data System (ADS)

Loong, Cheng Ning; Chang, C. C.; Dimitrakopoulos, Elias G.

2018-03-01

Mitigating excessive vibration of civil engineering structures using various types of devices has been a conspicuous research topic in the past few decades. Some devices, such as electromagnetic transducers, which have a capability of exerting control forces while simultaneously harvesting energy, have been proposed recently. These devices make possible a self-regenerative system that can semi-actively mitigate structural vibration without the need of external energy. Integrating mechanical, electrical components, and control algorithms, these devices open up a new research domain that needs to be addressed. In this study, the feasibility of using an actor-critic based reinforcement learning control algorithm for simultaneous vibration control and energy harvesting for a civil engineering structure is investigated. The actor-critic based reinforcement learning control algorithm is a real-time, model-free adaptive technique that can adjust the controller parameters based on observations and reward signals without knowing the system characteristics. It is suitable for the control of a partially known nonlinear system with uncertain parameters. The feasibility of implementing this algorithm on a building structure equipped with an electromagnetic damper will be investigated in this study. Issues related to the modelling of learning algorithm, initialization and convergence will be presented and discussed.

Using Fuzzy Logic for Performance Evaluation in Reinforcement Learning

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.; Khedkar, Pratap S.

1992-01-01

Current reinforcement learning algorithms require long training periods which generally limit their applicability to small size problems. A new architecture is described which uses fuzzy rules to initialize its two neural networks: a neural network for performance evaluation and another for action selection. This architecture is applied to control of dynamic systems and it is demonstrated that it is possible to start with an approximate prior knowledge and learn to refine it through experiments using reinforcement learning.

FMRQ-A Multiagent Reinforcement Learning Algorithm for Fully Cooperative Tasks.

PubMed

Zhang, Zhen; Zhao, Dongbin; Gao, Junwei; Wang, Dongqing; Dai, Yujie

2017-06-01

In this paper, we propose a multiagent reinforcement learning algorithm dealing with fully cooperative tasks. The algorithm is called frequency of the maximum reward Q-learning (FMRQ). FMRQ aims to achieve one of the optimal Nash equilibria so as to optimize the performance index in multiagent systems. The frequency of obtaining the highest global immediate reward instead of immediate reward is used as the reinforcement signal. With FMRQ each agent does not need the observation of the other agents' actions and only shares its state and reward at each step. We validate FMRQ through case studies of repeated games: four cases of two-player two-action and one case of three-player two-action. It is demonstrated that FMRQ can converge to one of the optimal Nash equilibria in these cases. Moreover, comparison experiments on tasks with multiple states and finite steps are conducted. One is box-pushing and the other one is distributed sensor network problem. Experimental results show that the proposed algorithm outperforms others with higher performance.

A parameter control method in reinforcement learning to rapidly follow unexpected environmental changes.

PubMed

Murakoshi, Kazushi; Mizuno, Junya

2004-11-01

In order to rapidly follow unexpected environmental changes, we propose a parameter control method in reinforcement learning that changes each of learning parameters in appropriate directions. We determine each appropriate direction on the basis of relationships between behaviors and neuromodulators by considering an emergency as a key word. Computer experiments show that the agents using our proposed method could rapidly respond to unexpected environmental changes, not depending on either two reinforcement learning algorithms (Q-learning and actor-critic (AC) architecture) or two learning problems (discontinuous and continuous state-action problems).

Flow Navigation by Smart Microswimmers via Reinforcement Learning

NASA Astrophysics Data System (ADS)

Colabrese, Simona; Biferale, Luca; Celani, Antonio; Gustavsson, Kristian

2017-11-01

We have numerically modeled active particles which are able to acquire some limited knowledge of the fluid environment from simple mechanical cues and exert a control on their preferred steering direction. We show that those swimmers can learn effective strategies just by experience, using a reinforcement learning algorithm. As an example, we focus on smart gravitactic swimmers. These are active particles whose task is to reach the highest altitude within some time horizon, exploiting the underlying flow whenever possible. The reinforcement learning algorithm allows particles to learn effective strategies even in difficult situations when, in the absence of control, they would end up being trapped by flow structures. These strategies are highly nontrivial and cannot be easily guessed in advance. This work paves the way towards the engineering of smart microswimmers that solve difficult navigation problems. ERC AdG NewTURB 339032.

Learning-based traffic signal control algorithms with neighborhood information sharing: An application for sustainable mobility

DOE Office of Scientific and Technical Information (OSTI.GOV)

Aziz, H. M. Abdul; Zhu, Feng; Ukkusuri, Satish V.

Here, this research applies R-Markov Average Reward Technique based reinforcement learning (RL) algorithm, namely RMART, for vehicular signal control problem leveraging information sharing among signal controllers in connected vehicle environment. We implemented the algorithm in a network of 18 signalized intersections and compare the performance of RMART with fixed, adaptive, and variants of the RL schemes. Results show significant improvement in system performance for RMART algorithm with information sharing over both traditional fixed signal timing plans and real time adaptive control schemes. Additionally, the comparison with reinforcement learning algorithms including Q learning and SARSA indicate that RMART performs better atmore » higher congestion levels. Further, a multi-reward structure is proposed that dynamically adjusts the reward function with varying congestion states at the intersection. Finally, the results from test networks show significant reduction in emissions (CO, CO 2, NO x, VOC, PM 10) when RL algorithms are implemented compared to fixed signal timings and adaptive schemes.« less

Novelty and Inductive Generalization in Human Reinforcement Learning

PubMed Central

Gershman, Samuel J.; Niv, Yael

2015-01-01

In reinforcement learning, a decision maker searching for the most rewarding option is often faced with the question: what is the value of an option that has never been tried before? One way to frame this question is as an inductive problem: how can I generalize my previous experience with one set of options to a novel option? We show how hierarchical Bayesian inference can be used to solve this problem, and describe an equivalence between the Bayesian model and temporal difference learning algorithms that have been proposed as models of reinforcement learning in humans and animals. According to our view, the search for the best option is guided by abstract knowledge about the relationships between different options in an environment, resulting in greater search efficiency compared to traditional reinforcement learning algorithms previously applied to human cognition. In two behavioral experiments, we test several predictions of our model, providing evidence that humans learn and exploit structured inductive knowledge to make predictions about novel options. In light of this model, we suggest a new interpretation of dopaminergic responses to novelty. PMID:25808176

"Notice of Violation of IEEE Publication Principles" Multiobjective Reinforcement Learning: A Comprehensive Overview.

PubMed

Liu, Chunming; Xu, Xin; Hu, Dewen

2013-04-29

Reinforcement learning is a powerful mechanism for enabling agents to learn in an unknown environment, and most reinforcement learning algorithms aim to maximize some numerical value, which represents only one long-term objective. However, multiple long-term objectives are exhibited in many real-world decision and control problems; therefore, recently, there has been growing interest in solving multiobjective reinforcement learning (MORL) problems with multiple conflicting objectives. The aim of this paper is to present a comprehensive overview of MORL. In this paper, the basic architecture, research topics, and naive solutions of MORL are introduced at first. Then, several representative MORL approaches and some important directions of recent research are reviewed. The relationships between MORL and other related research are also discussed, which include multiobjective optimization, hierarchical reinforcement learning, and multi-agent reinforcement learning. Finally, research challenges and open problems of MORL techniques are highlighted.

On the integration of reinforcement learning and approximate reasoning for control

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.

1991-01-01

The author discusses the importance of strengthening the knowledge representation characteristic of reinforcement learning techniques using methods such as approximate reasoning. The ARIC (approximate reasoning-based intelligent control) architecture is an example of such a hybrid approach in which the fuzzy control rules are modified (fine-tuned) using reinforcement learning. ARIC also demonstrates that it is possible to start with an approximately correct control knowledge base and learn to refine this knowledge through further experience. On the other hand, techniques such as the TD (temporal difference) algorithm and Q-learning establish stronger theoretical foundations for their use in adaptive control and also in stability analysis of hybrid reinforcement learning and approximate reasoning-based controllers.

Optimal and Autonomous Control Using Reinforcement Learning: A Survey.

PubMed

Kiumarsi, Bahare; Vamvoudakis, Kyriakos G; Modares, Hamidreza; Lewis, Frank L

2018-06-01

This paper reviews the current state of the art on reinforcement learning (RL)-based feedback control solutions to optimal regulation and tracking of single and multiagent systems. Existing RL solutions to both optimal and control problems, as well as graphical games, will be reviewed. RL methods learn the solution to optimal control and game problems online and using measured data along the system trajectories. We discuss Q-learning and the integral RL algorithm as core algorithms for discrete-time (DT) and continuous-time (CT) systems, respectively. Moreover, we discuss a new direction of off-policy RL for both CT and DT systems. Finally, we review several applications.

Robust reinforcement learning.

PubMed

Morimoto, Jun; Doya, Kenji

2005-02-01

This letter proposes a new reinforcement learning (RL) paradigm that explicitly takes into account input disturbance as well as modeling errors. The use of environmental models in RL is quite popular for both offline learning using simulations and for online action planning. However, the difference between the model and the real environment can lead to unpredictable, and often unwanted, results. Based on the theory of H(infinity) control, we consider a differential game in which a "disturbing" agent tries to make the worst possible disturbance while a "control" agent tries to make the best control input. The problem is formulated as finding a min-max solution of a value function that takes into account the amount of the reward and the norm of the disturbance. We derive online learning algorithms for estimating the value function and for calculating the worst disturbance and the best control in reference to the value function. We tested the paradigm, which we call robust reinforcement learning (RRL), on the control task of an inverted pendulum. In the linear domain, the policy and the value function learned by online algorithms coincided with those derived analytically by the linear H(infinity) control theory. For a fully nonlinear swing-up task, RRL achieved robust performance with changes in the pendulum weight and friction, while a standard reinforcement learning algorithm could not deal with these changes. We also applied RRL to the cart-pole swing-up task, and a robust swing-up policy was acquired.

Reinforcement learning improves behaviour from evaluative feedback

NASA Astrophysics Data System (ADS)

Littman, Michael L.

2015-05-01

Reinforcement learning is a branch of machine learning concerned with using experience gained through interacting with the world and evaluative feedback to improve a system's ability to make behavioural decisions. It has been called the artificial intelligence problem in a microcosm because learning algorithms must act autonomously to perform well and achieve their goals. Partly driven by the increasing availability of rich data, recent years have seen exciting advances in the theory and practice of reinforcement learning, including developments in fundamental technical areas such as generalization, planning, exploration and empirical methodology, leading to increasing applicability to real-life problems.

Reinforcement learning improves behaviour from evaluative feedback.

PubMed

Littman, Michael L

2015-05-28

Reinforcement learning is a branch of machine learning concerned with using experience gained through interacting with the world and evaluative feedback to improve a system's ability to make behavioural decisions. It has been called the artificial intelligence problem in a microcosm because learning algorithms must act autonomously to perform well and achieve their goals. Partly driven by the increasing availability of rich data, recent years have seen exciting advances in the theory and practice of reinforcement learning, including developments in fundamental technical areas such as generalization, planning, exploration and empirical methodology, leading to increasing applicability to real-life problems.

Flow Navigation by Smart Microswimmers via Reinforcement Learning

NASA Astrophysics Data System (ADS)

Colabrese, Simona; Gustavsson, Kristian; Celani, Antonio; Biferale, Luca

2017-04-01

Smart active particles can acquire some limited knowledge of the fluid environment from simple mechanical cues and exert a control on their preferred steering direction. Their goal is to learn the best way to navigate by exploiting the underlying flow whenever possible. As an example, we focus our attention on smart gravitactic swimmers. These are active particles whose task is to reach the highest altitude within some time horizon, given the constraints enforced by fluid mechanics. By means of numerical experiments, we show that swimmers indeed learn nearly optimal strategies just by experience. A reinforcement learning algorithm allows particles to learn effective strategies even in difficult situations when, in the absence of control, they would end up being trapped by flow structures. These strategies are highly nontrivial and cannot be easily guessed in advance. This Letter illustrates the potential of reinforcement learning algorithms to model adaptive behavior in complex flows and paves the way towards the engineering of smart microswimmers that solve difficult navigation problems.

The Effects of Observation of Learn Units during Reinforcement and Correction Conditions on the Rate of Learning Math Algorithms by Fifth Grade Students

ERIC Educational Resources Information Center

Neu, Jessica Adele

2013-01-01

I conducted two studies on the comparative effects of the observation of learn units during (a) reinforcement or (b) correction conditions on the acquisition of math objectives. The dependent variables were the within-session cumulative numbers of correct responses emitted during observational sessions. The independent variables were the…

Efficient model learning methods for actor-critic control.

PubMed

Grondman, Ivo; Vaandrager, Maarten; Buşoniu, Lucian; Babuska, Robert; Schuitema, Erik

2012-06-01

We propose two new actor-critic algorithms for reinforcement learning. Both algorithms use local linear regression (LLR) to learn approximations of the functions involved. A crucial feature of the algorithms is that they also learn a process model, and this, in combination with LLR, provides an efficient policy update for faster learning. The first algorithm uses a novel model-based update rule for the actor parameters. The second algorithm does not use an explicit actor but learns a reference model which represents a desired behavior, from which desired control actions can be calculated using the inverse of the learned process model. The two novel methods and a standard actor-critic algorithm are applied to the pendulum swing-up problem, in which the novel methods achieve faster learning than the standard algorithm.

Bio-robots automatic navigation with graded electric reward stimulation based on Reinforcement Learning.

PubMed

Zhang, Chen; Sun, Chao; Gao, Liqiang; Zheng, Nenggan; Chen, Weidong; Zheng, Xiaoxiang

2013-01-01

Bio-robots based on brain computer interface (BCI) suffer from the lack of considering the characteristic of the animals in navigation. This paper proposed a new method for bio-robots' automatic navigation combining the reward generating algorithm base on Reinforcement Learning (RL) with the learning intelligence of animals together. Given the graded electrical reward, the animal e.g. the rat, intends to seek the maximum reward while exploring an unknown environment. Since the rat has excellent spatial recognition, the rat-robot and the RL algorithm can convergent to an optimal route by co-learning. This work has significant inspiration for the practical development of bio-robots' navigation with hybrid intelligence.

Training Strategies for the M1 Abrams Tank Driver Trainer

DTIC Science & Technology

1984-10-01

positive reinforcement. The automatic freeze after incorrect performance, for example, may even be aversive to the trainee. The TECEP learning algorithms ...Aagard, J.A. and Braby, R. Learning Guidelines and Algorithms for Types of Training Objectives. (TAEG Report No. 23). Orlando, FL: Training Analysis and...checklist ite. flake it identical to operational setting. () Cresponde to the g;uideli ne number Tor thiss oast. Figure B-I. Learning Algorithm for

Feature Reinforcement Learning: Part I. Unstructured MDPs

NASA Astrophysics Data System (ADS)

Hutter, Marcus

2009-12-01

General-purpose, intelligent, learning agents cycle through sequences of observations, actions, and rewards that are complex, uncertain, unknown, and non-Markovian. On the other hand, reinforcement learning is well-developed for small finite state Markov decision processes (MDPs). Up to now, extracting the right state representations out of bare observations, that is, reducing the general agent setup to the MDP framework, is an art that involves significant effort by designers. The primary goal of this work is to automate the reduction process and thereby significantly expand the scope of many existing reinforcement learning algorithms and the agents that employ them. Before we can think of mechanizing this search for suitable MDPs, we need a formal objective criterion. The main contribution of this article is to develop such a criterion. I also integrate the various parts into one learning algorithm. Extensions to more realistic dynamic Bayesian networks are developed in Part II (Hutter, 2009c). The role of POMDPs is also considered there.

Utilising reinforcement learning to develop strategies for driving auditory neural implants.

PubMed

Lee, Geoffrey W; Zambetta, Fabio; Li, Xiaodong; Paolini, Antonio G

2016-08-01

In this paper we propose a novel application of reinforcement learning to the area of auditory neural stimulation. We aim to develop a simulation environment which is based off real neurological responses to auditory and electrical stimulation in the cochlear nucleus (CN) and inferior colliculus (IC) of an animal model. Using this simulator we implement closed loop reinforcement learning algorithms to determine which methods are most effective at learning effective acoustic neural stimulation strategies. By recording a comprehensive set of acoustic frequency presentations and neural responses from a set of animals we created a large database of neural responses to acoustic stimulation. Extensive electrical stimulation in the CN and the recording of neural responses in the IC provides a mapping of how the auditory system responds to electrical stimuli. The combined dataset is used as the foundation for the simulator, which is used to implement and test learning algorithms. Reinforcement learning, utilising a modified n-Armed Bandit solution, is implemented to demonstrate the model's function. We show the ability to effectively learn stimulation patterns which mimic the cochlea's ability to covert acoustic frequencies to neural activity. Time taken to learn effective replication using neural stimulation takes less than 20 min under continuous testing. These results show the utility of reinforcement learning in the field of neural stimulation. These results can be coupled with existing sound processing technologies to develop new auditory prosthetics that are adaptable to the recipients current auditory pathway. The same process can theoretically be abstracted to other sensory and motor systems to develop similar electrical replication of neural signals.

Multiagent Reinforcement Learning With Sparse Interactions by Negotiation and Knowledge Transfer.

PubMed

Zhou, Luowei; Yang, Pei; Chen, Chunlin; Gao, Yang

2017-05-01

Reinforcement learning has significant applications for multiagent systems, especially in unknown dynamic environments. However, most multiagent reinforcement learning (MARL) algorithms suffer from such problems as exponential computation complexity in the joint state-action space, which makes it difficult to scale up to realistic multiagent problems. In this paper, a novel algorithm named negotiation-based MARL with sparse interactions (NegoSIs) is presented. In contrast to traditional sparse-interaction-based MARL algorithms, NegoSI adopts the equilibrium concept and makes it possible for agents to select the nonstrict equilibrium-dominating strategy profile (nonstrict EDSP) or meta equilibrium for their joint actions. The presented NegoSI algorithm consists of four parts: 1) the equilibrium-based framework for sparse interactions; 2) the negotiation for the equilibrium set; 3) the minimum variance method for selecting one joint action; and 4) the knowledge transfer of local Q -values. In this integrated algorithm, three techniques, i.e., unshared value functions, equilibrium solutions, and sparse interactions are adopted to achieve privacy protection, better coordination and lower computational complexity, respectively. To evaluate the performance of the presented NegoSI algorithm, two groups of experiments are carried out regarding three criteria: 1) steps of each episode; 2) rewards of each episode; and 3) average runtime. The first group of experiments is conducted using six grid world games and shows fast convergence and high scalability of the presented algorithm. Then in the second group of experiments NegoSI is applied to an intelligent warehouse problem and simulated results demonstrate the effectiveness of the presented NegoSI algorithm compared with other state-of-the-art MARL algorithms.

Q-Learning-Based Adjustable Fixed-Phase Quantum Grover Search Algorithm

NASA Astrophysics Data System (ADS)

Guo, Ying; Shi, Wensha; Wang, Yijun; Hu, Jiankun

2017-02-01

We demonstrate that the rotation phase can be suitably chosen to increase the efficiency of the phase-based quantum search algorithm, leading to a dynamic balance between iterations and success probabilities of the fixed-phase quantum Grover search algorithm with Q-learning for a given number of solutions. In this search algorithm, the proposed Q-learning algorithm, which is a model-free reinforcement learning strategy in essence, is used for performing a matching algorithm based on the fraction of marked items λ and the rotation phase α. After establishing the policy function α = π(λ), we complete the fixed-phase Grover algorithm, where the phase parameter is selected via the learned policy. Simulation results show that the Q-learning-based Grover search algorithm (QLGA) enables fewer iterations and gives birth to higher success probabilities. Compared with the conventional Grover algorithms, it avoids the optimal local situations, thereby enabling success probabilities to approach one.

Batch Mode Reinforcement Learning based on the Synthesis of Artificial Trajectories

PubMed Central

Fonteneau, Raphael; Murphy, Susan A.; Wehenkel, Louis; Ernst, Damien

2013-01-01

In this paper, we consider the batch mode reinforcement learning setting, where the central problem is to learn from a sample of trajectories a policy that satisfies or optimizes a performance criterion. We focus on the continuous state space case for which usual resolution schemes rely on function approximators either to represent the underlying control problem or to represent its value function. As an alternative to the use of function approximators, we rely on the synthesis of “artificial trajectories” from the given sample of trajectories, and show that this idea opens new avenues for designing and analyzing algorithms for batch mode reinforcement learning. PMID:24049244

Changes in corticostriatal connectivity during reinforcement learning in humans.

PubMed

Horga, Guillermo; Maia, Tiago V; Marsh, Rachel; Hao, Xuejun; Xu, Dongrong; Duan, Yunsuo; Tau, Gregory Z; Graniello, Barbara; Wang, Zhishun; Kangarlu, Alayar; Martinez, Diana; Packard, Mark G; Peterson, Bradley S

2015-02-01

Many computational models assume that reinforcement learning relies on changes in synaptic efficacy between cortical regions representing stimuli and striatal regions involved in response selection, but this assumption has thus far lacked empirical support in humans. We recorded hemodynamic signals with fMRI while participants navigated a virtual maze to find hidden rewards. We fitted a reinforcement-learning algorithm to participants' choice behavior and evaluated the neural activity and the changes in functional connectivity related to trial-by-trial learning variables. Activity in the posterior putamen during choice periods increased progressively during learning. Furthermore, the functional connections between the sensorimotor cortex and the posterior putamen strengthened progressively as participants learned the task. These changes in corticostriatal connectivity differentiated participants who learned the task from those who did not. These findings provide a direct link between changes in corticostriatal connectivity and learning, thereby supporting a central assumption common to several computational models of reinforcement learning. © 2014 Wiley Periodicals, Inc.

A fuzzy reinforcement learning approach to power control in wireless transmitters.

PubMed

Vengerov, David; Bambos, Nicholas; Berenji, Hamid R

2005-08-01

We address the issue of power-controlled shared channel access in wireless networks supporting packetized data traffic. We formulate this problem using the dynamic programming framework and present a new distributed fuzzy reinforcement learning algorithm (ACFRL-2) capable of adequately solving a class of problems to which the power control problem belongs. Our experimental results show that the algorithm converges almost deterministically to a neighborhood of optimal parameter values, as opposed to a very noisy stochastic convergence of earlier algorithms. The main tradeoff facing a transmitter is to balance its current power level with future backlog in the presence of stochastically changing interference. Simulation experiments demonstrate that the ACFRL-2 algorithm achieves significant performance gains over the standard power control approach used in CDMA2000. Such a large improvement is explained by the fact that ACFRL-2 allows transmitters to learn implicit coordination policies, which back off under stressful channel conditions as opposed to engaging in escalating "power wars."

Constructing Temporally Extended Actions through Incremental Community Detection

PubMed Central

Li, Ge

2018-01-01

Hierarchical reinforcement learning works on temporally extended actions or skills to facilitate learning. How to automatically form such abstraction is challenging, and many efforts tackle this issue in the options framework. While various approaches exist to construct options from different perspectives, few of them concentrate on options' adaptability during learning. This paper presents an algorithm to create options and enhance their quality online. Both aspects operate on detected communities of the learning environment's state transition graph. We first construct options from initial samples as the basis of online learning. Then a rule-based community revision algorithm is proposed to update graph partitions, based on which existing options can be continuously tuned. Experimental results in two problems indicate that options from initial samples may perform poorly in more complex environments, and our presented strategy can effectively improve options and get better results compared with flat reinforcement learning. PMID:29849543

Off-Policy Integral Reinforcement Learning Method to Solve Nonlinear Continuous-Time Multiplayer Nonzero-Sum Games.

PubMed

Song, Ruizhuo; Lewis, Frank L; Wei, Qinglai

2017-03-01

This paper establishes an off-policy integral reinforcement learning (IRL) method to solve nonlinear continuous-time (CT) nonzero-sum (NZS) games with unknown system dynamics. The IRL algorithm is presented to obtain the iterative control and off-policy learning is used to allow the dynamics to be completely unknown. Off-policy IRL is designed to do policy evaluation and policy improvement in the policy iteration algorithm. Critic and action networks are used to obtain the performance index and control for each player. The gradient descent algorithm makes the update of critic and action weights simultaneously. The convergence analysis of the weights is given. The asymptotic stability of the closed-loop system and the existence of Nash equilibrium are proved. The simulation study demonstrates the effectiveness of the developed method for nonlinear CT NZS games with unknown system dynamics.

Self-Paced Prioritized Curriculum Learning With Coverage Penalty in Deep Reinforcement Learning.

PubMed

Ren, Zhipeng; Dong, Daoyi; Li, Huaxiong; Chen, Chunlin; Zhipeng Ren; Daoyi Dong; Huaxiong Li; Chunlin Chen; Dong, Daoyi; Li, Huaxiong; Chen, Chunlin; Ren, Zhipeng

2018-06-01

In this paper, a new training paradigm is proposed for deep reinforcement learning using self-paced prioritized curriculum learning with coverage penalty. The proposed deep curriculum reinforcement learning (DCRL) takes the most advantage of experience replay by adaptively selecting appropriate transitions from replay memory based on the complexity of each transition. The criteria of complexity in DCRL consist of self-paced priority as well as coverage penalty. The self-paced priority reflects the relationship between the temporal-difference error and the difficulty of the current curriculum for sample efficiency. The coverage penalty is taken into account for sample diversity. With comparison to deep Q network (DQN) and prioritized experience replay (PER) methods, the DCRL algorithm is evaluated on Atari 2600 games, and the experimental results show that DCRL outperforms DQN and PER on most of these games. More results further show that the proposed curriculum training paradigm of DCRL is also applicable and effective for other memory-based deep reinforcement learning approaches, such as double DQN and dueling network. All the experimental results demonstrate that DCRL can achieve improved training efficiency and robustness for deep reinforcement learning.

Benchmarking for Bayesian Reinforcement Learning

PubMed Central

Ernst, Damien; Couëtoux, Adrien

2016-01-01

In the Bayesian Reinforcement Learning (BRL) setting, agents try to maximise the collected rewards while interacting with their environment while using some prior knowledge that is accessed beforehand. Many BRL algorithms have already been proposed, but the benchmarks used to compare them are only relevant for specific cases. The paper addresses this problem, and provides a new BRL comparison methodology along with the corresponding open source library. In this methodology, a comparison criterion that measures the performance of algorithms on large sets of Markov Decision Processes (MDPs) drawn from some probability distributions is defined. In order to enable the comparison of non-anytime algorithms, our methodology also includes a detailed analysis of the computation time requirement of each algorithm. Our library is released with all source code and documentation: it includes three test problems, each of which has two different prior distributions, and seven state-of-the-art RL algorithms. Finally, our library is illustrated by comparing all the available algorithms and the results are discussed. PMID:27304891

Benchmarking for Bayesian Reinforcement Learning.

PubMed

Castronovo, Michael; Ernst, Damien; Couëtoux, Adrien; Fonteneau, Raphael

2016-01-01

In the Bayesian Reinforcement Learning (BRL) setting, agents try to maximise the collected rewards while interacting with their environment while using some prior knowledge that is accessed beforehand. Many BRL algorithms have already been proposed, but the benchmarks used to compare them are only relevant for specific cases. The paper addresses this problem, and provides a new BRL comparison methodology along with the corresponding open source library. In this methodology, a comparison criterion that measures the performance of algorithms on large sets of Markov Decision Processes (MDPs) drawn from some probability distributions is defined. In order to enable the comparison of non-anytime algorithms, our methodology also includes a detailed analysis of the computation time requirement of each algorithm. Our library is released with all source code and documentation: it includes three test problems, each of which has two different prior distributions, and seven state-of-the-art RL algorithms. Finally, our library is illustrated by comparing all the available algorithms and the results are discussed.

Machine learning in cardiovascular medicine: are we there yet?

PubMed

Shameer, Khader; Johnson, Kipp W; Glicksberg, Benjamin S; Dudley, Joel T; Sengupta, Partho P

2018-01-19

Artificial intelligence (AI) broadly refers to analytical algorithms that iteratively learn from data, allowing computers to find hidden insights without being explicitly programmed where to look. These include a family of operations encompassing several terms like machine learning, cognitive learning, deep learning and reinforcement learning-based methods that can be used to integrate and interpret complex biomedical and healthcare data in scenarios where traditional statistical methods may not be able to perform. In this review article, we discuss the basics of machine learning algorithms and what potential data sources exist; evaluate the need for machine learning; and examine the potential limitations and challenges of implementing machine in the context of cardiovascular medicine. The most promising avenues for AI in medicine are the development of automated risk prediction algorithms which can be used to guide clinical care; use of unsupervised learning techniques to more precisely phenotype complex disease; and the implementation of reinforcement learning algorithms to intelligently augment healthcare providers. The utility of a machine learning-based predictive model will depend on factors including data heterogeneity, data depth, data breadth, nature of modelling task, choice of machine learning and feature selection algorithms, and orthogonal evidence. A critical understanding of the strength and limitations of various methods and tasks amenable to machine learning is vital. By leveraging the growing corpus of big data in medicine, we detail pathways by which machine learning may facilitate optimal development of patient-specific models for improving diagnoses, intervention and outcome in cardiovascular medicine. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2018. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

Reinforcement learning: Solving two case studies

NASA Astrophysics Data System (ADS)

Duarte, Ana Filipa; Silva, Pedro; dos Santos, Cristina Peixoto

2012-09-01

Reinforcement Learning algorithms offer interesting features for the control of autonomous systems, such as the ability to learn from direct interaction with the environment, and the use of a simple reward signalas opposed to the input-outputs pairsused in classic supervised learning. The reward signal indicates the success of failure of the actions executed by the agent in the environment. In this work, are described RL algorithmsapplied to two case studies: the Crawler robot and the widely known inverted pendulum. We explore RL capabilities to autonomously learn a basic locomotion pattern in the Crawler, andapproach the balancing problem of biped locomotion using the inverted pendulum.

Decentralized learning in Markov games.

PubMed

Vrancx, Peter; Verbeeck, Katja; Nowé, Ann

2008-08-01

Learning automata (LA) were recently shown to be valuable tools for designing multiagent reinforcement learning algorithms. One of the principal contributions of the LA theory is that a set of decentralized independent LA is able to control a finite Markov chain with unknown transition probabilities and rewards. In this paper, we propose to extend this algorithm to Markov games--a straightforward extension of single-agent Markov decision problems to distributed multiagent decision problems. We show that under the same ergodic assumptions of the original theorem, the extended algorithm will converge to a pure equilibrium point between agent policies.

Reinforcement learning with Marr.

PubMed

Niv, Yael; Langdon, Angela

2016-10-01

To many, the poster child for David Marr's famous three levels of scientific inquiry is reinforcement learning-a computational theory of reward optimization, which readily prescribes algorithmic solutions that evidence striking resemblance to signals found in the brain, suggesting a straightforward neural implementation. Here we review questions that remain open at each level of analysis, concluding that the path forward to their resolution calls for inspiration across levels, rather than a focus on mutual constraints.

Safe Exploration Algorithms for Reinforcement Learning Controllers.

PubMed

Mannucci, Tommaso; van Kampen, Erik-Jan; de Visser, Cornelis; Chu, Qiping

2018-04-01

Self-learning approaches, such as reinforcement learning, offer new possibilities for autonomous control of uncertain or time-varying systems. However, exploring an unknown environment under limited prediction capabilities is a challenge for a learning agent. If the environment is dangerous, free exploration can result in physical damage or in an otherwise unacceptable behavior. With respect to existing methods, the main contribution of this paper is the definition of a new approach that does not require global safety functions, nor specific formulations of the dynamics or of the environment, but relies on interval estimation of the dynamics of the agent during the exploration phase, assuming a limited capability of the agent to perceive the presence of incoming fatal states. Two algorithms are presented with this approach. The first is the Safety Handling Exploration with Risk Perception Algorithm (SHERPA), which provides safety by individuating temporary safety functions, called backups. SHERPA is shown in a simulated, simplified quadrotor task, for which dangerous states are avoided. The second algorithm, denominated OptiSHERPA, can safely handle more dynamically complex systems for which SHERPA is not sufficient through the use of safety metrics. An application of OptiSHERPA is simulated on an aircraft altitude control task.

An Energy-Efficient Spectrum-Aware Reinforcement Learning-Based Clustering Algorithm for Cognitive Radio Sensor Networks

PubMed Central

Mustapha, Ibrahim; Ali, Borhanuddin Mohd; Rasid, Mohd Fadlee A.; Sali, Aduwati; Mohamad, Hafizal

2015-01-01

It is well-known that clustering partitions network into logical groups of nodes in order to achieve energy efficiency and to enhance dynamic channel access in cognitive radio through cooperative sensing. While the topic of energy efficiency has been well investigated in conventional wireless sensor networks, the latter has not been extensively explored. In this paper, we propose a reinforcement learning-based spectrum-aware clustering algorithm that allows a member node to learn the energy and cooperative sensing costs for neighboring clusters to achieve an optimal solution. Each member node selects an optimal cluster that satisfies pairwise constraints, minimizes network energy consumption and enhances channel sensing performance through an exploration technique. We first model the network energy consumption and then determine the optimal number of clusters for the network. The problem of selecting an optimal cluster is formulated as a Markov Decision Process (MDP) in the algorithm and the obtained simulation results show convergence, learning and adaptability of the algorithm to dynamic environment towards achieving an optimal solution. Performance comparisons of our algorithm with the Groupwise Spectrum Aware (GWSA)-based algorithm in terms of Sum of Square Error (SSE), complexity, network energy consumption and probability of detection indicate improved performance from the proposed approach. The results further reveal that an energy savings of 9% and a significant Primary User (PU) detection improvement can be achieved with the proposed approach. PMID:26287191

An Energy-Efficient Spectrum-Aware Reinforcement Learning-Based Clustering Algorithm for Cognitive Radio Sensor Networks.

PubMed

Mustapha, Ibrahim; Mohd Ali, Borhanuddin; Rasid, Mohd Fadlee A; Sali, Aduwati; Mohamad, Hafizal

2015-08-13

It is well-known that clustering partitions network into logical groups of nodes in order to achieve energy efficiency and to enhance dynamic channel access in cognitive radio through cooperative sensing. While the topic of energy efficiency has been well investigated in conventional wireless sensor networks, the latter has not been extensively explored. In this paper, we propose a reinforcement learning-based spectrum-aware clustering algorithm that allows a member node to learn the energy and cooperative sensing costs for neighboring clusters to achieve an optimal solution. Each member node selects an optimal cluster that satisfies pairwise constraints, minimizes network energy consumption and enhances channel sensing performance through an exploration technique. We first model the network energy consumption and then determine the optimal number of clusters for the network. The problem of selecting an optimal cluster is formulated as a Markov Decision Process (MDP) in the algorithm and the obtained simulation results show convergence, learning and adaptability of the algorithm to dynamic environment towards achieving an optimal solution. Performance comparisons of our algorithm with the Groupwise Spectrum Aware (GWSA)-based algorithm in terms of Sum of Square Error (SSE), complexity, network energy consumption and probability of detection indicate improved performance from the proposed approach. The results further reveal that an energy savings of 9% and a significant Primary User (PU) detection improvement can be achieved with the proposed approach.

GA-based fuzzy reinforcement learning for control of a magnetic bearing system.

PubMed

Lin, C T; Jou, C P

2000-01-01

This paper proposes a TD (temporal difference) and GA (genetic algorithm)-based reinforcement (TDGAR) learning method and applies it to the control of a real magnetic bearing system. The TDGAR learning scheme is a new hybrid GA, which integrates the TD prediction method and the GA to perform the reinforcement learning task. The TDGAR learning system is composed of two integrated feedforward networks. One neural network acts as a critic network to guide the learning of the other network (the action network) which determines the outputs (actions) of the TDGAR learning system. The action network can be a normal neural network or a neural fuzzy network. Using the TD prediction method, the critic network can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the action network. The action network uses the GA to adapt itself according to the internal reinforcement signal. The key concept of the TDGAR learning scheme is to formulate the internal reinforcement signal as the fitness function for the GA such that the GA can evaluate the candidate solutions (chromosomes) regularly, even during periods without external feedback from the environment. This enables the GA to proceed to new generations regularly without waiting for the arrival of the external reinforcement signal. This can usually accelerate the GA learning since a reinforcement signal may only be available at a time long after a sequence of actions has occurred in the reinforcement learning problem. The proposed TDGAR learning system has been used to control an active magnetic bearing (AMB) system in practice. A systematic design procedure is developed to achieve successful integration of all the subsystems including magnetic suspension, mechanical structure, and controller training. The results show that the TDGAR learning scheme can successfully find a neural controller or a neural fuzzy controller for a self-designed magnetic bearing system.

Quantum reinforcement learning.

PubMed

Dong, Daoyi; Chen, Chunlin; Li, Hanxiong; Tarn, Tzyh-Jong

2008-10-01

The key approaches for machine learning, particularly learning in unknown probabilistic environments, are new representations and computation mechanisms. In this paper, a novel quantum reinforcement learning (QRL) method is proposed by combining quantum theory and reinforcement learning (RL). Inspired by the state superposition principle and quantum parallelism, a framework of a value-updating algorithm is introduced. The state (action) in traditional RL is identified as the eigen state (eigen action) in QRL. The state (action) set can be represented with a quantum superposition state, and the eigen state (eigen action) can be obtained by randomly observing the simulated quantum state according to the collapse postulate of quantum measurement. The probability of the eigen action is determined by the probability amplitude, which is updated in parallel according to rewards. Some related characteristics of QRL such as convergence, optimality, and balancing between exploration and exploitation are also analyzed, which shows that this approach makes a good tradeoff between exploration and exploitation using the probability amplitude and can speedup learning through the quantum parallelism. To evaluate the performance and practicability of QRL, several simulated experiments are given, and the results demonstrate the effectiveness and superiority of the QRL algorithm for some complex problems. This paper is also an effective exploration on the application of quantum computation to artificial intelligence.

Learning Multirobot Hose Transportation and Deployment by Distributed Round-Robin Q-Learning.

PubMed

Fernandez-Gauna, Borja; Etxeberria-Agiriano, Ismael; Graña, Manuel

2015-01-01

Multi-Agent Reinforcement Learning (MARL) algorithms face two main difficulties: the curse of dimensionality, and environment non-stationarity due to the independent learning processes carried out by the agents concurrently. In this paper we formalize and prove the convergence of a Distributed Round Robin Q-learning (D-RR-QL) algorithm for cooperative systems. The computational complexity of this algorithm increases linearly with the number of agents. Moreover, it eliminates environment non sta tionarity by carrying a round-robin scheduling of the action selection and execution. That this learning scheme allows the implementation of Modular State-Action Vetoes (MSAV) in cooperative multi-agent systems, which speeds up learning convergence in over-constrained systems by vetoing state-action pairs which lead to undesired termination states (UTS) in the relevant state-action subspace. Each agent's local state-action value function learning is an independent process, including the MSAV policies. Coordination of locally optimal policies to obtain the global optimal joint policy is achieved by a greedy selection procedure using message passing. We show that D-RR-QL improves over state-of-the-art approaches, such as Distributed Q-Learning, Team Q-Learning and Coordinated Reinforcement Learning in a paradigmatic Linked Multi-Component Robotic System (L-MCRS) control problem: the hose transportation task. L-MCRS are over-constrained systems with many UTS induced by the interaction of the passive linking element and the active mobile robots.

Derivatives of logarithmic stationary distributions for policy gradient reinforcement learning.

PubMed

Morimura, Tetsuro; Uchibe, Eiji; Yoshimoto, Junichiro; Peters, Jan; Doya, Kenji

2010-02-01

Most conventional policy gradient reinforcement learning (PGRL) algorithms neglect (or do not explicitly make use of) a term in the average reward gradient with respect to the policy parameter. That term involves the derivative of the stationary state distribution that corresponds to the sensitivity of its distribution to changes in the policy parameter. Although the bias introduced by this omission can be reduced by setting the forgetting rate gamma for the value functions close to 1, these algorithms do not permit gamma to be set exactly at gamma = 1. In this article, we propose a method for estimating the log stationary state distribution derivative (LSD) as a useful form of the derivative of the stationary state distribution through backward Markov chain formulation and a temporal difference learning framework. A new policy gradient (PG) framework with an LSD is also proposed, in which the average reward gradient can be estimated by setting gamma = 0, so it becomes unnecessary to learn the value functions. We also test the performance of the proposed algorithms using simple benchmark tasks and show that these can improve the performances of existing PG methods.

A graph-based evolutionary algorithm: Genetic Network Programming (GNP) and its extension using reinforcement learning.

PubMed

Mabu, Shingo; Hirasawa, Kotaro; Hu, Jinglu

2007-01-01

This paper proposes a graph-based evolutionary algorithm called Genetic Network Programming (GNP). Our goal is to develop GNP, which can deal with dynamic environments efficiently and effectively, based on the distinguished expression ability of the graph (network) structure. The characteristics of GNP are as follows. 1) GNP programs are composed of a number of nodes which execute simple judgment/processing, and these nodes are connected by directed links to each other. 2) The graph structure enables GNP to re-use nodes, thus the structure can be very compact. 3) The node transition of GNP is executed according to its node connections without any terminal nodes, thus the past history of the node transition affects the current node to be used and this characteristic works as an implicit memory function. These structural characteristics are useful for dealing with dynamic environments. Furthermore, we propose an extended algorithm, "GNP with Reinforcement Learning (GNPRL)" which combines evolution and reinforcement learning in order to create effective graph structures and obtain better results in dynamic environments. In this paper, we applied GNP to the problem of determining agents' behavior to evaluate its effectiveness. Tileworld was used as the simulation environment. The results show some advantages for GNP over conventional methods.

Reinforcement and inference in cross-situational word learning.

PubMed

Tilles, Paulo F C; Fontanari, José F

2013-01-01

Cross-situational word learning is based on the notion that a learner can determine the referent of a word by finding something in common across many observed uses of that word. Here we propose an adaptive learning algorithm that contains a parameter that controls the strength of the reinforcement applied to associations between concurrent words and referents, and a parameter that regulates inference, which includes built-in biases, such as mutual exclusivity, and information of past learning events. By adjusting these parameters so that the model predictions agree with data from representative experiments on cross-situational word learning, we were able to explain the learning strategies adopted by the participants of those experiments in terms of a trade-off between reinforcement and inference. These strategies can vary wildly depending on the conditions of the experiments. For instance, for fast mapping experiments (i.e., the correct referent could, in principle, be inferred in a single observation) inference is prevalent, whereas for segregated contextual diversity experiments (i.e., the referents are separated in groups and are exhibited with members of their groups only) reinforcement is predominant. Other experiments are explained with more balanced doses of reinforcement and inference.

Reinforcement learning produces dominant strategies for the Iterated Prisoner's Dilemma.

PubMed

Harper, Marc; Knight, Vincent; Jones, Martin; Koutsovoulos, Georgios; Glynatsi, Nikoleta E; Campbell, Owen

2017-01-01

We present tournament results and several powerful strategies for the Iterated Prisoner's Dilemma created using reinforcement learning techniques (evolutionary and particle swarm algorithms). These strategies are trained to perform well against a corpus of over 170 distinct opponents, including many well-known and classic strategies. All the trained strategies win standard tournaments against the total collection of other opponents. The trained strategies and one particular human made designed strategy are the top performers in noisy tournaments also.

Autonomous learning based on cost assumptions: theoretical studies and experiments in robot control.

PubMed

Ribeiro, C H; Hemerly, E M

2000-02-01

Autonomous learning techniques are based on experience acquisition. In most realistic applications, experience is time-consuming: it implies sensor reading, actuator control and algorithmic update, constrained by the learning system dynamics. The information crudeness upon which classical learning algorithms operate make such problems too difficult and unrealistic. Nonetheless, additional information for facilitating the learning process ideally should be embedded in such a way that the structural, well-studied characteristics of these fundamental algorithms are maintained. We investigate in this article a more general formulation of the Q-learning method that allows for a spreading of information derived from single updates towards a neighbourhood of the instantly visited state and converges to optimality. We show how this new formulation can be used as a mechanism to safely embed prior knowledge about the structure of the state space, and demonstrate it in a modified implementation of a reinforcement learning algorithm in a real robot navigation task.

Improving Robot Motor Learning with Negatively Valenced Reinforcement Signals

PubMed Central

Navarro-Guerrero, Nicolás; Lowe, Robert J.; Wermter, Stefan

2017-01-01

Both nociception and punishment signals have been used in robotics. However, the potential for using these negatively valenced types of reinforcement learning signals for robot learning has not been exploited in detail yet. Nociceptive signals are primarily used as triggers of preprogrammed action sequences. Punishment signals are typically disembodied, i.e., with no or little relation to the agent-intrinsic limitations, and they are often used to impose behavioral constraints. Here, we provide an alternative approach for nociceptive signals as drivers of learning rather than simple triggers of preprogrammed behavior. Explicitly, we use nociception to expand the state space while we use punishment as a negative reinforcement learning signal. We compare the performance—in terms of task error, the amount of perceived nociception, and length of learned action sequences—of different neural networks imbued with punishment-based reinforcement signals for inverse kinematic learning. We contrast the performance of a version of the neural network that receives nociceptive inputs to that without such a process. Furthermore, we provide evidence that nociception can improve learning—making the algorithm more robust against network initializations—as well as behavioral performance by reducing the task error, perceived nociception, and length of learned action sequences. Moreover, we provide evidence that punishment, at least as typically used within reinforcement learning applications, may be detrimental in all relevant metrics. PMID:28420976

Multi-Objective Reinforcement Learning-based Deep Neural Networks for Cognitive Space Communications

NASA Technical Reports Server (NTRS)

Ferreria, Paulo; Paffenroth, Randy; Wyglinski, Alexander M.; Hackett, Timothy; Bilen, Sven; Reinhart, Richard; Mortensen, Dale

2017-01-01

Future communication subsystems of space exploration missions can potentially benefit from software-defined radios (SDRs) controlled by machine learning algorithms. In this paper, we propose a novel hybrid radio resource allocation management control algorithm that integrates multi-objective reinforcement learning and deep artificial neural networks. The objective is to efficiently manage communications system resources by monitoring performance functions with common dependent variables that result in conflicting goals. The uncertainty in the performance of thousands of different possible combinations of radio parameters makes the trade-off between exploration and exploitation in reinforcement learning (RL) much more challenging for future critical space-based missions. Thus, the system should spend as little time as possible on exploring actions, and whenever it explores an action, it should perform at acceptable levels most of the time. The proposed approach enables on-line learning by interactions with the environment and restricts poor resource allocation performance through virtual environment exploration. Improvements in the multiobjective performance can be achieved via transmitter parameter adaptation on a packet-basis, with poorly predicted performance promptly resulting in rejected decisions. Simulations presented in this work considered the DVB-S2 standard adaptive transmitter parameters and additional ones expected to be present in future adaptive radio systems. Performance results are provided by analysis of the proposed hybrid algorithm when operating across a satellite communication channel from Earth to GEO orbit during clear sky conditions. The proposed approach constitutes part of the core cognitive engine proof-of-concept to be delivered to the NASA Glenn Research Center SCaN Testbed located onboard the International Space Station.

Multi-Objective Reinforcement Learning-Based Deep Neural Networks for Cognitive Space Communications

NASA Technical Reports Server (NTRS)

Ferreria, Paulo Victor R.; Paffenroth, Randy; Wyglinski, Alexander M.; Hackett, Timothy M.; Bilen, Sven G.; Reinhart, Richard C.; Mortensen, Dale J.

2017-01-01

Future communication subsystems of space exploration missions can potentially benefit from software-defined radios (SDRs) controlled by machine learning algorithms. In this paper, we propose a novel hybrid radio resource allocation management control algorithm that integrates multi-objective reinforcement learning and deep artificial neural networks. The objective is to efficiently manage communications system resources by monitoring performance functions with common dependent variables that result in conflicting goals. The uncertainty in the performance of thousands of different possible combinations of radio parameters makes the trade-off between exploration and exploitation in reinforcement learning (RL) much more challenging for future critical space-based missions. Thus, the system should spend as little time as possible on exploring actions, and whenever it explores an action, it should perform at acceptable levels most of the time. The proposed approach enables on-line learning by interactions with the environment and restricts poor resource allocation performance through virtual environment exploration. Improvements in the multiobjective performance can be achieved via transmitter parameter adaptation on a packet-basis, with poorly predicted performance promptly resulting in rejected decisions. Simulations presented in this work considered the DVB-S2 standard adaptive transmitter parameters and additional ones expected to be present in future adaptive radio systems. Performance results are provided by analysis of the proposed hybrid algorithm when operating across a satellite communication channel from Earth to GEO orbit during clear sky conditions. The proposed approach constitutes part of the core cognitive engine proof-of-concept to be delivered to the NASA Glenn Research Center SCaN Testbed located onboard the International Space Station.

Reinforcement learning produces dominant strategies for the Iterated Prisoner’s Dilemma

PubMed Central

Jones, Martin; Koutsovoulos, Georgios; Glynatsi, Nikoleta E.; Campbell, Owen

2017-01-01

We present tournament results and several powerful strategies for the Iterated Prisoner’s Dilemma created using reinforcement learning techniques (evolutionary and particle swarm algorithms). These strategies are trained to perform well against a corpus of over 170 distinct opponents, including many well-known and classic strategies. All the trained strategies win standard tournaments against the total collection of other opponents. The trained strategies and one particular human made designed strategy are the top performers in noisy tournaments also. PMID:29228001

Synchronization of Chaotic Systems without Direct Connections Using Reinforcement Learning

NASA Astrophysics Data System (ADS)

Sato, Norihisa; Adachi, Masaharu

In this paper, we propose a control method for the synchronization of chaotic systems that does not require the systems to be connected, unlike existing methods such as that proposed by Pecora and Carroll in 1990. The method is based on the reinforcement learning algorithm. We apply our method to two discrete-time chaotic systems with mismatched parameters and achieve M step delay synchronization. Moreover, we extend the proposed method to the synchronization of continuous-time chaotic systems.

Genetic reinforcement learning through symbiotic evolution for fuzzy controller design.

PubMed

Juang, C F; Lin, J Y; Lin, C T

2000-01-01

An efficient genetic reinforcement learning algorithm for designing fuzzy controllers is proposed in this paper. The genetic algorithm (GA) adopted in this paper is based upon symbiotic evolution which, when applied to fuzzy controller design, complements the local mapping property of a fuzzy rule. Using this Symbiotic-Evolution-based Fuzzy Controller (SEFC) design method, the number of control trials, as well as consumed CPU time, are considerably reduced when compared to traditional GA-based fuzzy controller design methods and other types of genetic reinforcement learning schemes. Moreover, unlike traditional fuzzy controllers, which partition the input space into a grid, SEFC partitions the input space in a flexible way, thus creating fewer fuzzy rules. In SEFC, different types of fuzzy rules whose consequent parts are singletons, fuzzy sets, or linear equations (TSK-type fuzzy rules) are allowed. Further, the free parameters (e.g., centers and widths of membership functions) and fuzzy rules are all tuned automatically. For the TSK-type fuzzy rule especially, which put the proposed learning algorithm in use, only the significant input variables are selected to participate in the consequent of a rule. The proposed SEFC design method has been applied to different simulated control problems, including the cart-pole balancing system, a magnetic levitation system, and a water bath temperature control system. The proposed SEFC has been verified to be efficient and superior from these control problems, and from comparisons with some traditional GA-based fuzzy systems.

A spiking neural network model of model-free reinforcement learning with high-dimensional sensory input and perceptual ambiguity.

PubMed

Nakano, Takashi; Otsuka, Makoto; Yoshimoto, Junichiro; Doya, Kenji

2015-01-01

A theoretical framework of reinforcement learning plays an important role in understanding action selection in animals. Spiking neural networks provide a theoretically grounded means to test computational hypotheses on neurally plausible algorithms of reinforcement learning through numerical simulation. However, most of these models cannot handle observations which are noisy, or occurred in the past, even though these are inevitable and constraining features of learning in real environments. This class of problem is formally known as partially observable reinforcement learning (PORL) problems. It provides a generalization of reinforcement learning to partially observable domains. In addition, observations in the real world tend to be rich and high-dimensional. In this work, we use a spiking neural network model to approximate the free energy of a restricted Boltzmann machine and apply it to the solution of PORL problems with high-dimensional observations. Our spiking network model solves maze tasks with perceptually ambiguous high-dimensional observations without knowledge of the true environment. An extended model with working memory also solves history-dependent tasks. The way spiking neural networks handle PORL problems may provide a glimpse into the underlying laws of neural information processing which can only be discovered through such a top-down approach.

A Spiking Neural Network Model of Model-Free Reinforcement Learning with High-Dimensional Sensory Input and Perceptual Ambiguity

PubMed Central

Nakano, Takashi; Otsuka, Makoto; Yoshimoto, Junichiro; Doya, Kenji

2015-01-01

A theoretical framework of reinforcement learning plays an important role in understanding action selection in animals. Spiking neural networks provide a theoretically grounded means to test computational hypotheses on neurally plausible algorithms of reinforcement learning through numerical simulation. However, most of these models cannot handle observations which are noisy, or occurred in the past, even though these are inevitable and constraining features of learning in real environments. This class of problem is formally known as partially observable reinforcement learning (PORL) problems. It provides a generalization of reinforcement learning to partially observable domains. In addition, observations in the real world tend to be rich and high-dimensional. In this work, we use a spiking neural network model to approximate the free energy of a restricted Boltzmann machine and apply it to the solution of PORL problems with high-dimensional observations. Our spiking network model solves maze tasks with perceptually ambiguous high-dimensional observations without knowledge of the true environment. An extended model with working memory also solves history-dependent tasks. The way spiking neural networks handle PORL problems may provide a glimpse into the underlying laws of neural information processing which can only be discovered through such a top-down approach. PMID:25734662

Mastering the game of Go without human knowledge.

PubMed

Silver, David; Schrittwieser, Julian; Simonyan, Karen; Antonoglou, Ioannis; Huang, Aja; Guez, Arthur; Hubert, Thomas; Baker, Lucas; Lai, Matthew; Bolton, Adrian; Chen, Yutian; Lillicrap, Timothy; Hui, Fan; Sifre, Laurent; van den Driessche, George; Graepel, Thore; Hassabis, Demis

2017-10-18

A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks. These neural networks were trained by supervised learning from human expert moves, and by reinforcement learning from self-play. Here we introduce an algorithm based solely on reinforcement learning, without human data, guidance or domain knowledge beyond game rules. AlphaGo becomes its own teacher: a neural network is trained to predict AlphaGo's own move selections and also the winner of AlphaGo's games. This neural network improves the strength of the tree search, resulting in higher quality move selection and stronger self-play in the next iteration. Starting tabula rasa, our new program AlphaGo Zero achieved superhuman performance, winning 100-0 against the previously published, champion-defeating AlphaGo.

Mastering the game of Go without human knowledge

NASA Astrophysics Data System (ADS)

Silver, David; Schrittwieser, Julian; Simonyan, Karen; Antonoglou, Ioannis; Huang, Aja; Guez, Arthur; Hubert, Thomas; Baker, Lucas; Lai, Matthew; Bolton, Adrian; Chen, Yutian; Lillicrap, Timothy; Hui, Fan; Sifre, Laurent; van den Driessche, George; Graepel, Thore; Hassabis, Demis

2017-10-01

A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks. These neural networks were trained by supervised learning from human expert moves, and by reinforcement learning from self-play. Here we introduce an algorithm based solely on reinforcement learning, without human data, guidance or domain knowledge beyond game rules. AlphaGo becomes its own teacher: a neural network is trained to predict AlphaGo’s own move selections and also the winner of AlphaGo’s games. This neural network improves the strength of the tree search, resulting in higher quality move selection and stronger self-play in the next iteration. Starting tabula rasa, our new program AlphaGo Zero achieved superhuman performance, winning 100-0 against the previously published, champion-defeating AlphaGo.

Application of Reinforcement Learning in Cognitive Radio Networks: Models and Algorithms

PubMed Central

Yau, Kok-Lim Alvin; Poh, Geong-Sen; Chien, Su Fong; Al-Rawi, Hasan A. A.

2014-01-01

Cognitive radio (CR) enables unlicensed users to exploit the underutilized spectrum in licensed spectrum whilst minimizing interference to licensed users. Reinforcement learning (RL), which is an artificial intelligence approach, has been applied to enable each unlicensed user to observe and carry out optimal actions for performance enhancement in a wide range of schemes in CR, such as dynamic channel selection and channel sensing. This paper presents new discussions of RL in the context of CR networks. It provides an extensive review on how most schemes have been approached using the traditional and enhanced RL algorithms through state, action, and reward representations. Examples of the enhancements on RL, which do not appear in the traditional RL approach, are rules and cooperative learning. This paper also reviews performance enhancements brought about by the RL algorithms and open issues. This paper aims to establish a foundation in order to spark new research interests in this area. Our discussion has been presented in a tutorial manner so that it is comprehensive to readers outside the specialty of RL and CR. PMID:24995352

Fuzzy Sarsa with Focussed Replacing Eligibility Traces for Robust and Accurate Control

NASA Astrophysics Data System (ADS)

Kamdem, Sylvain; Ohki, Hidehiro; Sueda, Naomichi

Several methods of reinforcement learning in continuous state and action spaces that utilize fuzzy logic have been proposed in recent years. This paper introduces Fuzzy Sarsa(λ), an on-policy algorithm for fuzzy learning that relies on a novel way of computing replacing eligibility traces to accelerate the policy evaluation. It is tested against several temporal difference learning algorithms: Sarsa(λ), Fuzzy Q(λ), an earlier fuzzy version of Sarsa and an actor-critic algorithm. We perform detailed evaluations on two benchmark problems : a maze domain and the cart pole. Results of various tests highlight the strengths and weaknesses of these algorithms and show that Fuzzy Sarsa(λ) outperforms all other algorithms tested for a larger granularity of design and under noisy conditions. It is a highly competitive method of learning in realistic noisy domains where a denser fuzzy design over the state space is needed for a more precise control.

An incremental approach to genetic-algorithms-based classification.

PubMed

Guan, Sheng-Uei; Zhu, Fangming

2005-04-01

Incremental learning has been widely addressed in the machine learning literature to cope with learning tasks where the learning environment is ever changing or training samples become available over time. However, most research work explores incremental learning with statistical algorithms or neural networks, rather than evolutionary algorithms. The work in this paper employs genetic algorithms (GAs) as basic learning algorithms for incremental learning within one or more classifier agents in a multiagent environment. Four new approaches with different initialization schemes are proposed. They keep the old solutions and use an "integration" operation to integrate them with new elements to accommodate new attributes, while biased mutation and crossover operations are adopted to further evolve a reinforced solution. The simulation results on benchmark classification data sets show that the proposed approaches can deal with the arrival of new input attributes and integrate them with the original input space. It is also shown that the proposed approaches can be successfully used for incremental learning and improve classification rates as compared to the retraining GA. Possible applications for continuous incremental training and feature selection are also discussed.

Learning and tuning fuzzy logic controllers through reinforcements

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.; Khedkar, Pratap

1992-01-01

A new method for learning and tuning a fuzzy logic controller based on reinforcements from a dynamic system is presented. In particular, our Generalized Approximate Reasoning-based Intelligent Control (GARIC) architecture: (1) learns and tunes a fuzzy logic controller even when only weak reinforcements, such as a binary failure signal, is available; (2) introduces a new conjunction operator in computing the rule strengths of fuzzy control rules; (3) introduces a new localized mean of maximum (LMOM) method in combining the conclusions of several firing control rules; and (4) learns to produce real-valued control actions. Learning is achieved by integrating fuzzy inference into a feedforward network, which can then adaptively improve performance by using gradient descent methods. We extend the AHC algorithm of Barto, Sutton, and Anderson to include the prior control knowledge of human operators. The GARIC architecture is applied to a cart-pole balancing system and has demonstrated significant improvements in terms of the speed of learning and robustness to changes in the dynamic system's parameters over previous schemes for cart-pole balancing.

Implicit Value Updating Explains Transitive Inference Performance: The Betasort Model

PubMed Central

Jensen, Greg; Muñoz, Fabian; Alkan, Yelda; Ferrera, Vincent P.; Terrace, Herbert S.

2015-01-01

Transitive inference (the ability to infer that B > D given that B > C and C > D) is a widespread characteristic of serial learning, observed in dozens of species. Despite these robust behavioral effects, reinforcement learning models reliant on reward prediction error or associative strength routinely fail to perform these inferences. We propose an algorithm called betasort, inspired by cognitive processes, which performs transitive inference at low computational cost. This is accomplished by (1) representing stimulus positions along a unit span using beta distributions, (2) treating positive and negative feedback asymmetrically, and (3) updating the position of every stimulus during every trial, whether that stimulus was visible or not. Performance was compared for rhesus macaques, humans, and the betasort algorithm, as well as Q-learning, an established reward-prediction error (RPE) model. Of these, only Q-learning failed to respond above chance during critical test trials. Betasort’s success (when compared to RPE models) and its computational efficiency (when compared to full Markov decision process implementations) suggests that the study of reinforcement learning in organisms will be best served by a feature-driven approach to comparing formal models. PMID:26407227

Implicit Value Updating Explains Transitive Inference Performance: The Betasort Model.

PubMed

Jensen, Greg; Muñoz, Fabian; Alkan, Yelda; Ferrera, Vincent P; Terrace, Herbert S

2015-01-01

Transitive inference (the ability to infer that B > D given that B > C and C > D) is a widespread characteristic of serial learning, observed in dozens of species. Despite these robust behavioral effects, reinforcement learning models reliant on reward prediction error or associative strength routinely fail to perform these inferences. We propose an algorithm called betasort, inspired by cognitive processes, which performs transitive inference at low computational cost. This is accomplished by (1) representing stimulus positions along a unit span using beta distributions, (2) treating positive and negative feedback asymmetrically, and (3) updating the position of every stimulus during every trial, whether that stimulus was visible or not. Performance was compared for rhesus macaques, humans, and the betasort algorithm, as well as Q-learning, an established reward-prediction error (RPE) model. Of these, only Q-learning failed to respond above chance during critical test trials. Betasort's success (when compared to RPE models) and its computational efficiency (when compared to full Markov decision process implementations) suggests that the study of reinforcement learning in organisms will be best served by a feature-driven approach to comparing formal models.

Accelerating Multiagent Reinforcement Learning by Equilibrium Transfer.

PubMed

Hu, Yujing; Gao, Yang; An, Bo

2015-07-01

An important approach in multiagent reinforcement learning (MARL) is equilibrium-based MARL, which adopts equilibrium solution concepts in game theory and requires agents to play equilibrium strategies at each state. However, most existing equilibrium-based MARL algorithms cannot scale due to a large number of computationally expensive equilibrium computations (e.g., computing Nash equilibria is PPAD-hard) during learning. For the first time, this paper finds that during the learning process of equilibrium-based MARL, the one-shot games corresponding to each state's successive visits often have the same or similar equilibria (for some states more than 90% of games corresponding to successive visits have similar equilibria). Inspired by this observation, this paper proposes to use equilibrium transfer to accelerate equilibrium-based MARL. The key idea of equilibrium transfer is to reuse previously computed equilibria when each agent has a small incentive to deviate. By introducing transfer loss and transfer condition, a novel framework called equilibrium transfer-based MARL is proposed. We prove that although equilibrium transfer brings transfer loss, equilibrium-based MARL algorithms can still converge to an equilibrium policy under certain assumptions. Experimental results in widely used benchmarks (e.g., grid world game, soccer game, and wall game) show that the proposed framework: 1) not only significantly accelerates equilibrium-based MARL (up to 96.7% reduction in learning time), but also achieves higher average rewards than algorithms without equilibrium transfer and 2) scales significantly better than algorithms without equilibrium transfer when the state/action space grows and the number of agents increases.

Reinforcement learning agents providing advice in complex video games

NASA Astrophysics Data System (ADS)

Taylor, Matthew E.; Carboni, Nicholas; Fachantidis, Anestis; Vlahavas, Ioannis; Torrey, Lisa

2014-01-01

This article introduces a teacher-student framework for reinforcement learning, synthesising and extending material that appeared in conference proceedings [Torrey, L., & Taylor, M. E. (2013)]. Teaching on a budget: Agents advising agents in reinforcement learning. {Proceedings of the international conference on autonomous agents and multiagent systems}] and in a non-archival workshop paper [Carboni, N., &Taylor, M. E. (2013, May)]. Preliminary results for 1 vs. 1 tactics in StarCraft. {Proceedings of the adaptive and learning agents workshop (at AAMAS-13)}]. In this framework, a teacher agent instructs a student agent by suggesting actions the student should take as it learns. However, the teacher may only give such advice a limited number of times. We present several novel algorithms that teachers can use to budget their advice effectively, and we evaluate them in two complex video games: StarCraft and Pac-Man. Our results show that the same amount of advice, given at different moments, can have different effects on student learning, and that teachers can significantly affect student learning even when students use different learning methods and state representations.

Needs, Pains, and Motivations in Autonomous Agents.

PubMed

Starzyk, Janusz A; Graham, James; Puzio, Leszek

This paper presents the development of a motivated learning (ML) agent with symbolic I/O. Our earlier work on the ML agent was enhanced, giving it autonomy for interaction with other agents. Specifically, we equipped the agent with drives and pains that establish its motivations to learn how to respond to desired and undesired events and create related abstract goals. The purpose of this paper is to explore the autonomous development of motivations and memory in agents within a simulated environment. The ML agent has been implemented in a virtual environment created within the NeoAxis game engine. Additionally, to illustrate the benefits of an ML-based agent, we compared the performance of our algorithm against various reinforcement learning (RL) algorithms in a dynamic test scenario, and demonstrated that our ML agent learns better than any of the tested RL agents.This paper presents the development of a motivated learning (ML) agent with symbolic I/O. Our earlier work on the ML agent was enhanced, giving it autonomy for interaction with other agents. Specifically, we equipped the agent with drives and pains that establish its motivations to learn how to respond to desired and undesired events and create related abstract goals. The purpose of this paper is to explore the autonomous development of motivations and memory in agents within a simulated environment. The ML agent has been implemented in a virtual environment created within the NeoAxis game engine. Additionally, to illustrate the benefits of an ML-based agent, we compared the performance of our algorithm against various reinforcement learning (RL) algorithms in a dynamic test scenario, and demonstrated that our ML agent learns better than any of the tested RL agents.

A Discussion of Possibility of Reinforcement Learning Using Event-Related Potential in BCI

NASA Astrophysics Data System (ADS)

Yamagishi, Yuya; Tsubone, Tadashi; Wada, Yasuhiro

Recently, Brain computer interface (BCI) which is a direct connecting pathway an external device such as a computer or a robot and a human brain have gotten a lot of attention. Since BCI can control the machines as robots by using the brain activity without using the voluntary muscle, the BCI may become a useful communication tool for handicapped persons, for instance, amyotrophic lateral sclerosis patients. However, in order to realize the BCI system which can perform precise tasks on various environments, it is necessary to design the control rules to adapt to the dynamic environments. Reinforcement learning is one approach of the design of the control rule. If this reinforcement leaning can be performed by the brain activity, it leads to the attainment of BCI that has general versatility. In this research, we paid attention to P300 of event-related potential as an alternative signal of the reward of reinforcement learning. We discriminated between the success and the failure trials from P300 of the EEG of the single trial by using the proposed discrimination algorithm based on Support vector machine. The possibility of reinforcement learning was examined from the viewpoint of the number of discriminated trials. It was shown that there was a possibility to be able to learn in most subjects.

Distribution majorization of corner points by reinforcement learning for moving object detection

NASA Astrophysics Data System (ADS)

Wu, Hao; Yu, Hao; Zhou, Dongxiang; Cheng, Yongqiang

2018-04-01

Corner points play an important role in moving object detection, especially in the case of free-moving camera. Corner points provide more accurate information than other pixels and reduce the computation which is unnecessary. Previous works only use intensity information to locate the corner points, however, the information that former and the last frames provided also can be used. We utilize the information to focus on more valuable area and ignore the invaluable area. The proposed algorithm is based on reinforcement learning, which regards the detection of corner points as a Markov process. In the Markov model, the video to be detected is regarded as environment, the selections of blocks for one corner point are regarded as actions and the performance of detection is regarded as state. Corner points are assigned to be the blocks which are seperated from original whole image. Experimentally, we select a conventional method which uses marching and Random Sample Consensus algorithm to obtain objects as the main framework and utilize our algorithm to improve the result. The comparison between the conventional method and the same one with our algorithm show that our algorithm reduce 70% of the false detection.

Reinforcement active learning in the vibrissae system: optimal object localization.

PubMed

Gordon, Goren; Dorfman, Nimrod; Ahissar, Ehud

2013-01-01

Rats move their whiskers to acquire information about their environment. It has been observed that they palpate novel objects and objects they are required to localize in space. We analyze whisker-based object localization using two complementary paradigms, namely, active learning and intrinsic-reward reinforcement learning. Active learning algorithms select the next training samples according to the hypothesized solution in order to better discriminate between correct and incorrect labels. Intrinsic-reward reinforcement learning uses prediction errors as the reward to an actor-critic design, such that behavior converges to the one that optimizes the learning process. We show that in the context of object localization, the two paradigms result in palpation whisking as their respective optimal solution. These results suggest that rats may employ principles of active learning and/or intrinsic reward in tactile exploration and can guide future research to seek the underlying neuronal mechanisms that implement them. Furthermore, these paradigms are easily transferable to biomimetic whisker-based artificial sensors and can improve the active exploration of their environment. Copyright © 2012 Elsevier Ltd. All rights reserved.

Learning to use working memory: a reinforcement learning gating model of rule acquisition in rats

PubMed Central

Lloyd, Kevin; Becker, Nadine; Jones, Matthew W.; Bogacz, Rafal

2012-01-01

Learning to form appropriate, task-relevant working memory representations is a complex process central to cognition. Gating models frame working memory as a collection of past observations and use reinforcement learning (RL) to solve the problem of when to update these observations. Investigation of how gating models relate to brain and behavior remains, however, at an early stage. The current study sought to explore the ability of simple RL gating models to replicate rule learning behavior in rats. Rats were trained in a maze-based spatial learning task that required animals to make trial-by-trial choices contingent upon their previous experience. Using an abstract version of this task, we tested the ability of two gating algorithms, one based on the Actor-Critic and the other on the State-Action-Reward-State-Action (SARSA) algorithm, to generate behavior consistent with the rats'. Both models produced rule-acquisition behavior consistent with the experimental data, though only the SARSA gating model mirrored faster learning following rule reversal. We also found that both gating models learned multiple strategies in solving the initial task, a property which highlights the multi-agent nature of such models and which is of importance in considering the neural basis of individual differences in behavior. PMID:23115551

Tunnel Ventilation Control Using Reinforcement Learning Methodology

NASA Astrophysics Data System (ADS)

Chu, Baeksuk; Kim, Dongnam; Hong, Daehie; Park, Jooyoung; Chung, Jin Taek; Kim, Tae-Hyung

The main purpose of tunnel ventilation system is to maintain CO pollutant concentration and VI (visibility index) under an adequate level to provide drivers with comfortable and safe driving environment. Moreover, it is necessary to minimize power consumption used to operate ventilation system. To achieve the objectives, the control algorithm used in this research is reinforcement learning (RL) method. RL is a goal-directed learning of a mapping from situations to actions without relying on exemplary supervision or complete models of the environment. The goal of RL is to maximize a reward which is an evaluative feedback from the environment. In the process of constructing the reward of the tunnel ventilation system, two objectives listed above are included, that is, maintaining an adequate level of pollutants and minimizing power consumption. RL algorithm based on actor-critic architecture and gradient-following algorithm is adopted to the tunnel ventilation system. The simulations results performed with real data collected from existing tunnel ventilation system and real experimental verification are provided in this paper. It is confirmed that with the suggested controller, the pollutant level inside the tunnel was well maintained under allowable limit and the performance of energy consumption was improved compared to conventional control scheme.

Novelty and Inductive Generalization in Human Reinforcement Learning.

PubMed

Gershman, Samuel J; Niv, Yael

2015-07-01

In reinforcement learning (RL), a decision maker searching for the most rewarding option is often faced with the question: What is the value of an option that has never been tried before? One way to frame this question is as an inductive problem: How can I generalize my previous experience with one set of options to a novel option? We show how hierarchical Bayesian inference can be used to solve this problem, and we describe an equivalence between the Bayesian model and temporal difference learning algorithms that have been proposed as models of RL in humans and animals. According to our view, the search for the best option is guided by abstract knowledge about the relationships between different options in an environment, resulting in greater search efficiency compared to traditional RL algorithms previously applied to human cognition. In two behavioral experiments, we test several predictions of our model, providing evidence that humans learn and exploit structured inductive knowledge to make predictions about novel options. In light of this model, we suggest a new interpretation of dopaminergic responses to novelty. Copyright © 2015 Cognitive Science Society, Inc.

Spectrum Access In Cognitive Radio Using a Two-Stage Reinforcement Learning Approach

NASA Astrophysics Data System (ADS)

Raj, Vishnu; Dias, Irene; Tholeti, Thulasi; Kalyani, Sheetal

2018-02-01

With the advent of the 5th generation of wireless standards and an increasing demand for higher throughput, methods to improve the spectral efficiency of wireless systems have become very important. In the context of cognitive radio, a substantial increase in throughput is possible if the secondary user can make smart decisions regarding which channel to sense and when or how often to sense. Here, we propose an algorithm to not only select a channel for data transmission but also to predict how long the channel will remain unoccupied so that the time spent on channel sensing can be minimized. Our algorithm learns in two stages - a reinforcement learning approach for channel selection and a Bayesian approach to determine the optimal duration for which sensing can be skipped. Comparisons with other learning methods are provided through extensive simulations. We show that the number of sensing is minimized with negligible increase in primary interference; this implies that lesser energy is spent by the secondary user in sensing and also higher throughput is achieved by saving on sensing.

Reinforcement interval type-2 fuzzy controller design by online rule generation and q-value-aided ant colony optimization.

PubMed

Juang, Chia-Feng; Hsu, Chia-Hung

2009-12-01

This paper proposes a new reinforcement-learning method using online rule generation and Q-value-aided ant colony optimization (ORGQACO) for fuzzy controller design. The fuzzy controller is based on an interval type-2 fuzzy system (IT2FS). The antecedent part in the designed IT2FS uses interval type-2 fuzzy sets to improve controller robustness to noise. There are initially no fuzzy rules in the IT2FS. The ORGQACO concurrently designs both the structure and parameters of an IT2FS. We propose an online interval type-2 rule generation method for the evolution of system structure and flexible partitioning of the input space. Consequent part parameters in an IT2FS are designed using Q -values and the reinforcement local-global ant colony optimization algorithm. This algorithm selects the consequent part from a set of candidate actions according to ant pheromone trails and Q-values, both of which are updated using reinforcement signals. The ORGQACO design method is applied to the following three control problems: 1) truck-backing control; 2) magnetic-levitation control; and 3) chaotic-system control. The ORGQACO is compared with other reinforcement-learning methods to verify its efficiency and effectiveness. Comparisons with type-1 fuzzy systems verify the noise robustness property of using an IT2FS.

Application of fuzzy logic-neural network based reinforcement learning to proximity and docking operations: Translational controller results

NASA Technical Reports Server (NTRS)

Jani, Yashvant

1992-01-01

The reinforcement learning techniques developed at Ames Research Center are being applied to proximity and docking operations using the Shuttle and Solar Maximum Mission (SMM) satellite simulation. In utilizing these fuzzy learning techniques, we also use the Approximate Reasoning based Intelligent Control (ARIC) architecture, and so we use two terms interchangeable to imply the same. This activity is carried out in the Software Technology Laboratory utilizing the Orbital Operations Simulator (OOS). This report is the deliverable D3 in our project activity and provides the test results of the fuzzy learning translational controller. This report is organized in six sections. Based on our experience and analysis with the attitude controller, we have modified the basic configuration of the reinforcement learning algorithm in ARIC as described in section 2. The shuttle translational controller and its implementation in fuzzy learning architecture is described in section 3. Two test cases that we have performed are described in section 4. Our results and conclusions are discussed in section 5, and section 6 provides future plans and summary for the project.

Learning and tuning fuzzy logic controllers through reinforcements

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.; Khedkar, Pratap

1992-01-01

This paper presents a new method for learning and tuning a fuzzy logic controller based on reinforcements from a dynamic system. In particular, our generalized approximate reasoning-based intelligent control (GARIC) architecture (1) learns and tunes a fuzzy logic controller even when only weak reinforcement, such as a binary failure signal, is available; (2) introduces a new conjunction operator in computing the rule strengths of fuzzy control rules; (3) introduces a new localized mean of maximum (LMOM) method in combining the conclusions of several firing control rules; and (4) learns to produce real-valued control actions. Learning is achieved by integrating fuzzy inference into a feedforward neural network, which can then adaptively improve performance by using gradient descent methods. We extend the AHC algorithm of Barto et al. (1983) to include the prior control knowledge of human operators. The GARIC architecture is applied to a cart-pole balancing system and demonstrates significant improvements in terms of the speed of learning and robustness to changes in the dynamic system's parameters over previous schemes for cart-pole balancing.

Simple artificial neural networks that match probability and exploit and explore when confronting a multiarmed bandit.

PubMed

Dawson, Michael R W; Dupuis, Brian; Spetch, Marcia L; Kelly, Debbie M

2009-08-01

The matching law (Herrnstein 1961) states that response rates become proportional to reinforcement rates; this is related to the empirical phenomenon called probability matching (Vulkan 2000). Here, we show that a simple artificial neural network generates responses consistent with probability matching. This behavior was then used to create an operant procedure for network learning. We use the multiarmed bandit (Gittins 1989), a classic problem of choice behavior, to illustrate that operant training balances exploiting the bandit arm expected to pay off most frequently with exploring other arms. Perceptrons provide a medium for relating results from neural networks, genetic algorithms, animal learning, contingency theory, reinforcement learning, and theories of choice.

Reinforcement learning techniques for controlling resources in power networks

NASA Astrophysics Data System (ADS)

Kowli, Anupama Sunil

As power grids transition towards increased reliance on renewable generation, energy storage and demand response resources, an effective control architecture is required to harness the full functionalities of these resources. There is a critical need for control techniques that recognize the unique characteristics of the different resources and exploit the flexibility afforded by them to provide ancillary services to the grid. The work presented in this dissertation addresses these needs. Specifically, new algorithms are proposed, which allow control synthesis in settings wherein the precise distribution of the uncertainty and its temporal statistics are not known. These algorithms are based on recent developments in Markov decision theory, approximate dynamic programming and reinforcement learning. They impose minimal assumptions on the system model and allow the control to be "learned" based on the actual dynamics of the system. Furthermore, they can accommodate complex constraints such as capacity and ramping limits on generation resources, state-of-charge constraints on storage resources, comfort-related limitations on demand response resources and power flow limits on transmission lines. Numerical studies demonstrating applications of these algorithms to practical control problems in power systems are discussed. Results demonstrate how the proposed control algorithms can be used to improve the performance and reduce the computational complexity of the economic dispatch mechanism in a power network. We argue that the proposed algorithms are eminently suitable to develop operational decision-making tools for large power grids with many resources and many sources of uncertainty.

Joint Extraction of Entities and Relations Using Reinforcement Learning and Deep Learning.

PubMed

Feng, Yuntian; Zhang, Hongjun; Hao, Wenning; Chen, Gang

2017-01-01

We use both reinforcement learning and deep learning to simultaneously extract entities and relations from unstructured texts. For reinforcement learning, we model the task as a two-step decision process. Deep learning is used to automatically capture the most important information from unstructured texts, which represent the state in the decision process. By designing the reward function per step, our proposed method can pass the information of entity extraction to relation extraction and obtain feedback in order to extract entities and relations simultaneously. Firstly, we use bidirectional LSTM to model the context information, which realizes preliminary entity extraction. On the basis of the extraction results, attention based method can represent the sentences that include target entity pair to generate the initial state in the decision process. Then we use Tree-LSTM to represent relation mentions to generate the transition state in the decision process. Finally, we employ Q -Learning algorithm to get control policy π in the two-step decision process. Experiments on ACE2005 demonstrate that our method attains better performance than the state-of-the-art method and gets a 2.4% increase in recall-score.

Joint Extraction of Entities and Relations Using Reinforcement Learning and Deep Learning

PubMed Central

Zhang, Hongjun; Chen, Gang

2017-01-01

We use both reinforcement learning and deep learning to simultaneously extract entities and relations from unstructured texts. For reinforcement learning, we model the task as a two-step decision process. Deep learning is used to automatically capture the most important information from unstructured texts, which represent the state in the decision process. By designing the reward function per step, our proposed method can pass the information of entity extraction to relation extraction and obtain feedback in order to extract entities and relations simultaneously. Firstly, we use bidirectional LSTM to model the context information, which realizes preliminary entity extraction. On the basis of the extraction results, attention based method can represent the sentences that include target entity pair to generate the initial state in the decision process. Then we use Tree-LSTM to represent relation mentions to generate the transition state in the decision process. Finally, we employ Q-Learning algorithm to get control policy π in the two-step decision process. Experiments on ACE2005 demonstrate that our method attains better performance than the state-of-the-art method and gets a 2.4% increase in recall-score. PMID:28894463

Efficient Actor-Critic Algorithm with Hierarchical Model Learning and Planning

PubMed Central

Fu, QiMing

2016-01-01

To improve the convergence rate and the sample efficiency, two efficient learning methods AC-HMLP and RAC-HMLP (AC-HMLP with ℓ 2-regularization) are proposed by combining actor-critic algorithm with hierarchical model learning and planning. The hierarchical models consisting of the local and the global models, which are learned at the same time during learning of the value function and the policy, are approximated by local linear regression (LLR) and linear function approximation (LFA), respectively. Both the local model and the global model are applied to generate samples for planning; the former is used only if the state-prediction error does not surpass the threshold at each time step, while the latter is utilized at the end of each episode. The purpose of taking both models is to improve the sample efficiency and accelerate the convergence rate of the whole algorithm through fully utilizing the local and global information. Experimentally, AC-HMLP and RAC-HMLP are compared with three representative algorithms on two Reinforcement Learning (RL) benchmark problems. The results demonstrate that they perform best in terms of convergence rate and sample efficiency. PMID:27795704

Efficient Actor-Critic Algorithm with Hierarchical Model Learning and Planning.

PubMed

Zhong, Shan; Liu, Quan; Fu, QiMing

2016-01-01

To improve the convergence rate and the sample efficiency, two efficient learning methods AC-HMLP and RAC-HMLP (AC-HMLP with ℓ 2 -regularization) are proposed by combining actor-critic algorithm with hierarchical model learning and planning. The hierarchical models consisting of the local and the global models, which are learned at the same time during learning of the value function and the policy, are approximated by local linear regression (LLR) and linear function approximation (LFA), respectively. Both the local model and the global model are applied to generate samples for planning; the former is used only if the state-prediction error does not surpass the threshold at each time step, while the latter is utilized at the end of each episode. The purpose of taking both models is to improve the sample efficiency and accelerate the convergence rate of the whole algorithm through fully utilizing the local and global information. Experimentally, AC-HMLP and RAC-HMLP are compared with three representative algorithms on two Reinforcement Learning (RL) benchmark problems. The results demonstrate that they perform best in terms of convergence rate and sample efficiency.

Hierarchical extreme learning machine based reinforcement learning for goal localization

NASA Astrophysics Data System (ADS)

AlDahoul, Nouar; Zaw Htike, Zaw; Akmeliawati, Rini

2017-03-01

The objective of goal localization is to find the location of goals in noisy environments. Simple actions are performed to move the agent towards the goal. The goal detector should be capable of minimizing the error between the predicted locations and the true ones. Few regions need to be processed by the agent to reduce the computational effort and increase the speed of convergence. In this paper, reinforcement learning (RL) method was utilized to find optimal series of actions to localize the goal region. The visual data, a set of images, is high dimensional unstructured data and needs to be represented efficiently to get a robust detector. Different deep Reinforcement models have already been used to localize a goal but most of them take long time to learn the model. This long learning time results from the weights fine tuning stage that is applied iteratively to find an accurate model. Hierarchical Extreme Learning Machine (H-ELM) was used as a fast deep model that doesn’t fine tune the weights. In other words, hidden weights are generated randomly and output weights are calculated analytically. H-ELM algorithm was used in this work to find good features for effective representation. This paper proposes a combination of Hierarchical Extreme learning machine and Reinforcement learning to find an optimal policy directly from visual input. This combination outperforms other methods in terms of accuracy and learning speed. The simulations and results were analysed by using MATLAB.

Learning New Basic Movements for Robotics

NASA Astrophysics Data System (ADS)

Kober, Jens; Peters, Jan

Obtaining novel skills is one of the most important problems in robotics. Machine learning techniques may be a promising approach for automatic and autonomous acquisition of movement policies. However, this requires both an appropriate policy representation and suitable learning algorithms. Employing the most recent form of the dynamical systems motor primitives originally introduced by Ijspeert et al. [1], we show how both discrete and rhythmic tasks can be learned using a concerted approach of both imitation and reinforcement learning, and present our current best performing learning algorithms. Finally, we show that it is possible to include a start-up phase in rhythmic primitives. We apply our approach to two elementary movements, i.e., Ball-in-a-Cup and Ball-Paddling, which can be learned on a real Barrett WAM robot arm at a pace similar to human learning.

Discrete-Time Stable Generalized Self-Learning Optimal Control With Approximation Errors.

PubMed

Wei, Qinglai; Li, Benkai; Song, Ruizhuo

2018-04-01

In this paper, a generalized policy iteration (GPI) algorithm with approximation errors is developed for solving infinite horizon optimal control problems for nonlinear systems. The developed stable GPI algorithm provides a general structure of discrete-time iterative adaptive dynamic programming algorithms, by which most of the discrete-time reinforcement learning algorithms can be described using the GPI structure. It is for the first time that approximation errors are explicitly considered in the GPI algorithm. The properties of the stable GPI algorithm with approximation errors are analyzed. The admissibility of the approximate iterative control law can be guaranteed if the approximation errors satisfy the admissibility criteria. The convergence of the developed algorithm is established, which shows that the iterative value function is convergent to a finite neighborhood of the optimal performance index function, if the approximate errors satisfy the convergence criterion. Finally, numerical examples and comparisons are presented.

Multiagent cooperation and competition with deep reinforcement learning.

PubMed

Tampuu, Ardi; Matiisen, Tambet; Kodelja, Dorian; Kuzovkin, Ilya; Korjus, Kristjan; Aru, Juhan; Aru, Jaan; Vicente, Raul

2017-01-01

Evolution of cooperation and competition can appear when multiple adaptive agents share a biological, social, or technological niche. In the present work we study how cooperation and competition emerge between autonomous agents that learn by reinforcement while using only their raw visual input as the state representation. In particular, we extend the Deep Q-Learning framework to multiagent environments to investigate the interaction between two learning agents in the well-known video game Pong. By manipulating the classical rewarding scheme of Pong we show how competitive and collaborative behaviors emerge. We also describe the progression from competitive to collaborative behavior when the incentive to cooperate is increased. Finally we show how learning by playing against another adaptive agent, instead of against a hard-wired algorithm, results in more robust strategies. The present work shows that Deep Q-Networks can become a useful tool for studying decentralized learning of multiagent systems coping with high-dimensional environments.

A self-learning rule base for command following in dynamical systems

NASA Technical Reports Server (NTRS)

Tsai, Wei K.; Lee, Hon-Mun; Parlos, Alexander

1992-01-01

In this paper, a self-learning Rule Base for command following in dynamical systems is presented. The learning is accomplished though reinforcement learning using an associative memory called SAM. The main advantage of SAM is that it is a function approximator with explicit storage of training samples. A learning algorithm patterned after the dynamic programming is proposed. Two artificially created, unstable dynamical systems are used for testing, and the Rule Base was used to generate a feedback control to improve the command following ability of the otherwise uncontrolled systems. The numerical results are very encouraging. The controlled systems exhibit a more stable behavior and a better capability to follow reference commands. The rules resulting from the reinforcement learning are explicitly stored and they can be modified or augmented by human experts. Due to overlapping storage scheme of SAM, the stored rules are similar to fuzzy rules.

Multiagent cooperation and competition with deep reinforcement learning

PubMed Central

Kodelja, Dorian; Kuzovkin, Ilya; Korjus, Kristjan; Aru, Juhan; Aru, Jaan; Vicente, Raul

2017-01-01

Evolution of cooperation and competition can appear when multiple adaptive agents share a biological, social, or technological niche. In the present work we study how cooperation and competition emerge between autonomous agents that learn by reinforcement while using only their raw visual input as the state representation. In particular, we extend the Deep Q-Learning framework to multiagent environments to investigate the interaction between two learning agents in the well-known video game Pong. By manipulating the classical rewarding scheme of Pong we show how competitive and collaborative behaviors emerge. We also describe the progression from competitive to collaborative behavior when the incentive to cooperate is increased. Finally we show how learning by playing against another adaptive agent, instead of against a hard-wired algorithm, results in more robust strategies. The present work shows that Deep Q-Networks can become a useful tool for studying decentralized learning of multiagent systems coping with high-dimensional environments. PMID:28380078

Predictive representations can link model-based reinforcement learning to model-free mechanisms.

PubMed

Russek, Evan M; Momennejad, Ida; Botvinick, Matthew M; Gershman, Samuel J; Daw, Nathaniel D

2017-09-01

Humans and animals are capable of evaluating actions by considering their long-run future rewards through a process described using model-based reinforcement learning (RL) algorithms. The mechanisms by which neural circuits perform the computations prescribed by model-based RL remain largely unknown; however, multiple lines of evidence suggest that neural circuits supporting model-based behavior are structurally homologous to and overlapping with those thought to carry out model-free temporal difference (TD) learning. Here, we lay out a family of approaches by which model-based computation may be built upon a core of TD learning. The foundation of this framework is the successor representation, a predictive state representation that, when combined with TD learning of value predictions, can produce a subset of the behaviors associated with model-based learning, while requiring less decision-time computation than dynamic programming. Using simulations, we delineate the precise behavioral capabilities enabled by evaluating actions using this approach, and compare them to those demonstrated by biological organisms. We then introduce two new algorithms that build upon the successor representation while progressively mitigating its limitations. Because this framework can account for the full range of observed putatively model-based behaviors while still utilizing a core TD framework, we suggest that it represents a neurally plausible family of mechanisms for model-based evaluation.

Predictive representations can link model-based reinforcement learning to model-free mechanisms

PubMed Central

Botvinick, Matthew M.

2017-01-01

Humans and animals are capable of evaluating actions by considering their long-run future rewards through a process described using model-based reinforcement learning (RL) algorithms. The mechanisms by which neural circuits perform the computations prescribed by model-based RL remain largely unknown; however, multiple lines of evidence suggest that neural circuits supporting model-based behavior are structurally homologous to and overlapping with those thought to carry out model-free temporal difference (TD) learning. Here, we lay out a family of approaches by which model-based computation may be built upon a core of TD learning. The foundation of this framework is the successor representation, a predictive state representation that, when combined with TD learning of value predictions, can produce a subset of the behaviors associated with model-based learning, while requiring less decision-time computation than dynamic programming. Using simulations, we delineate the precise behavioral capabilities enabled by evaluating actions using this approach, and compare them to those demonstrated by biological organisms. We then introduce two new algorithms that build upon the successor representation while progressively mitigating its limitations. Because this framework can account for the full range of observed putatively model-based behaviors while still utilizing a core TD framework, we suggest that it represents a neurally plausible family of mechanisms for model-based evaluation. PMID:28945743

Kernel-based least squares policy iteration for reinforcement learning.

PubMed

Xu, Xin; Hu, Dewen; Lu, Xicheng

2007-07-01

In this paper, we present a kernel-based least squares policy iteration (KLSPI) algorithm for reinforcement learning (RL) in large or continuous state spaces, which can be used to realize adaptive feedback control of uncertain dynamic systems. By using KLSPI, near-optimal control policies can be obtained without much a priori knowledge on dynamic models of control plants. In KLSPI, Mercer kernels are used in the policy evaluation of a policy iteration process, where a new kernel-based least squares temporal-difference algorithm called KLSTD-Q is proposed for efficient policy evaluation. To keep the sparsity and improve the generalization ability of KLSTD-Q solutions, a kernel sparsification procedure based on approximate linear dependency (ALD) is performed. Compared to the previous works on approximate RL methods, KLSPI makes two progresses to eliminate the main difficulties of existing results. One is the better convergence and (near) optimality guarantee by using the KLSTD-Q algorithm for policy evaluation with high precision. The other is the automatic feature selection using the ALD-based kernel sparsification. Therefore, the KLSPI algorithm provides a general RL method with generalization performance and convergence guarantee for large-scale Markov decision problems (MDPs). Experimental results on a typical RL task for a stochastic chain problem demonstrate that KLSPI can consistently achieve better learning efficiency and policy quality than the previous least squares policy iteration (LSPI) algorithm. Furthermore, the KLSPI method was also evaluated on two nonlinear feedback control problems, including a ship heading control problem and the swing up control of a double-link underactuated pendulum called acrobot. Simulation results illustrate that the proposed method can optimize controller performance using little a priori information of uncertain dynamic systems. It is also demonstrated that KLSPI can be applied to online learning control by incorporating an initial controller to ensure online performance.

Investigation of Drive-Reinforcement Learning and Application of Learning to Flight Control

DTIC Science & Technology

1993-08-01

Attachment 1 138 Reprint of: Baird, L. (1991). Learning and Adaptive Hybrid Systems for Nonlinear Control, CSDL Report T-1099, M.S. Thesis , Department of...Aircraft, CSDL Report T-1127, S.M. Thesis , Department of Aeronautics and Astronautics, M.I.T. Attachment 3 351 . iprint of: Atkins, S. (1993...Incremental Synthesis of Optimal Control Laws Using Learning Algorithms, CSDL Report T-1181, S.M. Thesis , Department of Aeronautics and Astronautics, M.I.T

Monocular Depth Perception and Robotic Grasping of Novel Objects

DTIC Science & Technology

2009-06-01

resulting algorithm is able to learn monocular vision cues that accurately estimate the relative depths of obstacles in a scene. Reinforcement learning ... learning still make sense in these settings? Since many of the cues that are useful for estimating depth can be re-created in synthetic images, we...supervised learning approach to this problem, and use a Markov Random Field (MRF) to model the scene depth as a function of the image features. We show

Neural correlates of strategic reasoning during competitive games.

PubMed

Seo, Hyojung; Cai, Xinying; Donahue, Christopher H; Lee, Daeyeol

2014-10-17

Although human and animal behaviors are largely shaped by reinforcement and punishment, choices in social settings are also influenced by information about the knowledge and experience of other decision-makers. During competitive games, monkeys increased their payoffs by systematically deviating from a simple heuristic learning algorithm and thereby countering the predictable exploitation by their computer opponent. Neurons in the dorsomedial prefrontal cortex (dmPFC) signaled the animal's recent choice and reward history that reflected the computer's exploitative strategy. The strength of switching signals in the dmPFC also correlated with the animal's tendency to deviate from the heuristic learning algorithm. Therefore, the dmPFC might provide control signals for overriding simple heuristic learning algorithms based on the inferred strategies of the opponent. Copyright © 2014, American Association for the Advancement of Science.

An adaptive deep Q-learning strategy for handwritten digit recognition.

PubMed

Qiao, Junfei; Wang, Gongming; Li, Wenjing; Chen, Min

2018-02-22

Handwritten digits recognition is a challenging problem in recent years. Although many deep learning-based classification algorithms are studied for handwritten digits recognition, the recognition accuracy and running time still need to be further improved. In this paper, an adaptive deep Q-learning strategy is proposed to improve accuracy and shorten running time for handwritten digit recognition. The adaptive deep Q-learning strategy combines the feature-extracting capability of deep learning and the decision-making of reinforcement learning to form an adaptive Q-learning deep belief network (Q-ADBN). First, Q-ADBN extracts the features of original images using an adaptive deep auto-encoder (ADAE), and the extracted features are considered as the current states of Q-learning algorithm. Second, Q-ADBN receives Q-function (reward signal) during recognition of the current states, and the final handwritten digits recognition is implemented by maximizing the Q-function using Q-learning algorithm. Finally, experimental results from the well-known MNIST dataset show that the proposed Q-ADBN has a superiority to other similar methods in terms of accuracy and running time. Copyright © 2018 Elsevier Ltd. All rights reserved.

Reinforcement learning algorithms for robotic navigation in dynamic environments.

PubMed

Yen, Gary G; Hickey, Travis W

2004-04-01

The purpose of this study was to examine improvements to reinforcement learning (RL) algorithms in order to successfully interact within dynamic environments. The scope of the research was that of RL algorithms as applied to robotic navigation. Proposed improvements include: addition of a forgetting mechanism, use of feature based state inputs, and hierarchical structuring of an RL agent. Simulations were performed to evaluate the individual merits and flaws of each proposal, to compare proposed methods to prior established methods, and to compare proposed methods to theoretically optimal solutions. Incorporation of a forgetting mechanism did considerably improve the learning times of RL agents in a dynamic environment. However, direct implementation of a feature-based RL agent did not result in any performance enhancements, as pure feature-based navigation results in a lack of positional awareness, and the inability of the agent to determine the location of the goal state. Inclusion of a hierarchical structure in an RL agent resulted in significantly improved performance, specifically when one layer of the hierarchy included a feature-based agent for obstacle avoidance, and a standard RL agent for global navigation. In summary, the inclusion of a forgetting mechanism, and the use of a hierarchically structured RL agent offer substantially increased performance when compared to traditional RL agents navigating in a dynamic environment.

The Computational Development of Reinforcement Learning during Adolescence

PubMed Central

Palminteri, Stefano; Coricelli, Giorgio; Blakemore, Sarah-Jayne

2016-01-01

Adolescence is a period of life characterised by changes in learning and decision-making. Learning and decision-making do not rely on a unitary system, but instead require the coordination of different cognitive processes that can be mathematically formalised as dissociable computational modules. Here, we aimed to trace the developmental time-course of the computational modules responsible for learning from reward or punishment, and learning from counterfactual feedback. Adolescents and adults carried out a novel reinforcement learning paradigm in which participants learned the association between cues and probabilistic outcomes, where the outcomes differed in valence (reward versus punishment) and feedback was either partial or complete (either the outcome of the chosen option only, or the outcomes of both the chosen and unchosen option, were displayed). Computational strategies changed during development: whereas adolescents’ behaviour was better explained by a basic reinforcement learning algorithm, adults’ behaviour integrated increasingly complex computational features, namely a counterfactual learning module (enabling enhanced performance in the presence of complete feedback) and a value contextualisation module (enabling symmetrical reward and punishment learning). Unlike adults, adolescent performance did not benefit from counterfactual (complete) feedback. In addition, while adults learned symmetrically from both reward and punishment, adolescents learned from reward but were less likely to learn from punishment. This tendency to rely on rewards and not to consider alternative consequences of actions might contribute to our understanding of decision-making in adolescence. PMID:27322574

Towards autonomous neuroprosthetic control using Hebbian reinforcement learning.

PubMed

Mahmoudi, Babak; Pohlmeyer, Eric A; Prins, Noeline W; Geng, Shijia; Sanchez, Justin C

2013-12-01

Our goal was to design an adaptive neuroprosthetic controller that could learn the mapping from neural states to prosthetic actions and automatically adjust adaptation using only a binary evaluative feedback as a measure of desirability/undesirability of performance. Hebbian reinforcement learning (HRL) in a connectionist network was used for the design of the adaptive controller. The method combines the efficiency of supervised learning with the generality of reinforcement learning. The convergence properties of this approach were studied using both closed-loop control simulations and open-loop simulations that used primate neural data from robot-assisted reaching tasks. The HRL controller was able to perform classification and regression tasks using its episodic and sequential learning modes, respectively. In our experiments, the HRL controller quickly achieved convergence to an effective control policy, followed by robust performance. The controller also automatically stopped adapting the parameters after converging to a satisfactory control policy. Additionally, when the input neural vector was reorganized, the controller resumed adaptation to maintain performance. By estimating an evaluative feedback directly from the user, the HRL control algorithm may provide an efficient method for autonomous adaptation of neuroprosthetic systems. This method may enable the user to teach the controller the desired behavior using only a simple feedback signal.

Reinforcement learning for resource allocation in LEO satellite networks.

PubMed

Usaha, Wipawee; Barria, Javier A

2007-06-01

In this paper, we develop and assess online decision-making algorithms for call admission and routing for low Earth orbit (LEO) satellite networks. It has been shown in a recent paper that, in a LEO satellite system, a semi-Markov decision process formulation of the call admission and routing problem can achieve better performance in terms of an average revenue function than existing routing methods. However, the conventional dynamic programming (DP) numerical solution becomes prohibited as the problem size increases. In this paper, two solution methods based on reinforcement learning (RL) are proposed in order to circumvent the computational burden of DP. The first method is based on an actor-critic method with temporal-difference (TD) learning. The second method is based on a critic-only method, called optimistic TD learning. The algorithms enhance performance in terms of requirements in storage, computational complexity and computational time, and in terms of an overall long-term average revenue function that penalizes blocked calls. Numerical studies are carried out, and the results obtained show that the RL framework can achieve up to 56% higher average revenue over existing routing methods used in LEO satellite networks with reasonable storage and computational requirements.

Interconnected growing self-organizing maps for auditory and semantic acquisition modeling.

PubMed

Cao, Mengxue; Li, Aijun; Fang, Qiang; Kaufmann, Emily; Kröger, Bernd J

2014-01-01

Based on the incremental nature of knowledge acquisition, in this study we propose a growing self-organizing neural network approach for modeling the acquisition of auditory and semantic categories. We introduce an Interconnected Growing Self-Organizing Maps (I-GSOM) algorithm, which takes associations between auditory information and semantic information into consideration, in this paper. Direct phonetic-semantic association is simulated in order to model the language acquisition in early phases, such as the babbling and imitation stages, in which no phonological representations exist. Based on the I-GSOM algorithm, we conducted experiments using paired acoustic and semantic training data. We use a cyclical reinforcing and reviewing training procedure to model the teaching and learning process between children and their communication partners. A reinforcing-by-link training procedure and a link-forgetting procedure are introduced to model the acquisition of associative relations between auditory and semantic information. Experimental results indicate that (1) I-GSOM has good ability to learn auditory and semantic categories presented within the training data; (2) clear auditory and semantic boundaries can be found in the network representation; (3) cyclical reinforcing and reviewing training leads to a detailed categorization as well as to a detailed clustering, while keeping the clusters that have already been learned and the network structure that has already been developed stable; and (4) reinforcing-by-link training leads to well-perceived auditory-semantic associations. Our I-GSOM model suggests that it is important to associate auditory information with semantic information during language acquisition. Despite its high level of abstraction, our I-GSOM approach can be interpreted as a biologically-inspired neurocomputational model.

Model-Free control performance improvement using virtual reference feedback tuning and reinforcement Q-learning

NASA Astrophysics Data System (ADS)

Radac, Mircea-Bogdan; Precup, Radu-Emil; Roman, Raul-Cristian

2017-04-01

This paper proposes the combination of two model-free controller tuning techniques, namely linear virtual reference feedback tuning (VRFT) and nonlinear state-feedback Q-learning, referred to as a new mixed VRFT-Q learning approach. VRFT is first used to find stabilising feedback controller using input-output experimental data from the process in a model reference tracking setting. Reinforcement Q-learning is next applied in the same setting using input-state experimental data collected under perturbed VRFT to ensure good exploration. The Q-learning controller learned with a batch fitted Q iteration algorithm uses two neural networks, one for the Q-function estimator and one for the controller, respectively. The VRFT-Q learning approach is validated on position control of a two-degrees-of-motion open-loop stable multi input-multi output (MIMO) aerodynamic system (AS). Extensive simulations for the two independent control channels of the MIMO AS show that the Q-learning controllers clearly improve performance over the VRFT controllers.

Social learning through prediction error in the brain

NASA Astrophysics Data System (ADS)

Joiner, Jessica; Piva, Matthew; Turrin, Courtney; Chang, Steve W. C.

2017-06-01

Learning about the world is critical to survival and success. In social animals, learning about others is a necessary component of navigating the social world, ultimately contributing to increasing evolutionary fitness. How humans and nonhuman animals represent the internal states and experiences of others has long been a subject of intense interest in the developmental psychology tradition, and, more recently, in studies of learning and decision making involving self and other. In this review, we explore how psychology conceptualizes the process of representing others, and how neuroscience has uncovered correlates of reinforcement learning signals to explore the neural mechanisms underlying social learning from the perspective of representing reward-related information about self and other. In particular, we discuss self-referenced and other-referenced types of reward prediction errors across multiple brain structures that effectively allow reinforcement learning algorithms to mediate social learning. Prediction-based computational principles in the brain may be strikingly conserved between self-referenced and other-referenced information.

A Machine Learning Concept for DTN Routing

NASA Technical Reports Server (NTRS)

Dudukovich, Rachel; Hylton, Alan; Papachristou, Christos

2017-01-01

This paper discusses the concept and architecture of a machine learning based router for delay tolerant space networks. The techniques of reinforcement learning and Bayesian learning are used to supplement the routing decisions of the popular Contact Graph Routing algorithm. An introduction to the concepts of Contact Graph Routing, Q-routing and Naive Bayes classification are given. The development of an architecture for a cross-layer feedback framework for DTN (Delay-Tolerant Networking) protocols is discussed. Finally, initial simulation setup and results are given.

Reinforcement learning and decision making in monkeys during a competitive game.

PubMed

Lee, Daeyeol; Conroy, Michelle L; McGreevy, Benjamin P; Barraclough, Dominic J

2004-12-01

Animals living in a dynamic environment must adjust their decision-making strategies through experience. To gain insights into the neural basis of such adaptive decision-making processes, we trained monkeys to play a competitive game against a computer in an oculomotor free-choice task. The animal selected one of two visual targets in each trial and was rewarded only when it selected the same target as the computer opponent. To determine how the animal's decision-making strategy can be affected by the opponent's strategy, the computer opponent was programmed with three different algorithms that exploited different aspects of the animal's choice and reward history. When the computer selected its targets randomly with equal probabilities, animals selected one of the targets more often, violating the prediction of probability matching, and their choices were systematically influenced by the choice history of the two players. When the computer exploited only the animal's choice history but not its reward history, animal's choice became more independent of its own choice history but was still related to the choice history of the opponent. This bias was substantially reduced, but not completely eliminated, when the computer used the choice history of both players in making its predictions. These biases were consistent with the predictions of reinforcement learning, suggesting that the animals sought optimal decision-making strategies using reinforcement learning algorithms.

Fault-tolerant optimised tracking control for unknown discrete-time linear systems using a combined reinforcement learning and residual compensation methodology

NASA Astrophysics Data System (ADS)

Han, Ke-Zhen; Feng, Jian; Cui, Xiaohong

2017-10-01

This paper considers the fault-tolerant optimised tracking control (FTOTC) problem for unknown discrete-time linear system. A research scheme is proposed on the basis of data-based parity space identification, reinforcement learning and residual compensation techniques. The main characteristic of this research scheme lies in the parity-space-identification-based simultaneous tracking control and residual compensation. The specific technical line consists of four main contents: apply subspace aided method to design observer-based residual generator; use reinforcement Q-learning approach to solve optimised tracking control policy; rely on robust H∞ theory to achieve noise attenuation; adopt fault estimation triggered by residual generator to perform fault compensation. To clarify the design and implementation procedures, an integrated algorithm is further constructed to link up these four functional units. The detailed analysis and proof are subsequently given to explain the guaranteed FTOTC performance of the proposed conclusions. Finally, a case simulation is provided to verify its effectiveness.

Implementation of real-time energy management strategy based on reinforcement learning for hybrid electric vehicles and simulation validation

PubMed Central

Kong, Zehui; Liu, Teng

2017-01-01

To further improve the fuel economy of series hybrid electric tracked vehicles, a reinforcement learning (RL)-based real-time energy management strategy is developed in this paper. In order to utilize the statistical characteristics of online driving schedule effectively, a recursive algorithm for the transition probability matrix (TPM) of power-request is derived. The reinforcement learning (RL) is applied to calculate and update the control policy at regular time, adapting to the varying driving conditions. A facing-forward powertrain model is built in detail, including the engine-generator model, battery model and vehicle dynamical model. The robustness and adaptability of real-time energy management strategy are validated through the comparison with the stationary control strategy based on initial transition probability matrix (TPM) generated from a long naturalistic driving cycle in the simulation. Results indicate that proposed method has better fuel economy than stationary one and is more effective in real-time control. PMID:28671967

Implementation of real-time energy management strategy based on reinforcement learning for hybrid electric vehicles and simulation validation.

PubMed

Kong, Zehui; Zou, Yuan; Liu, Teng

2017-01-01

To further improve the fuel economy of series hybrid electric tracked vehicles, a reinforcement learning (RL)-based real-time energy management strategy is developed in this paper. In order to utilize the statistical characteristics of online driving schedule effectively, a recursive algorithm for the transition probability matrix (TPM) of power-request is derived. The reinforcement learning (RL) is applied to calculate and update the control policy at regular time, adapting to the varying driving conditions. A facing-forward powertrain model is built in detail, including the engine-generator model, battery model and vehicle dynamical model. The robustness and adaptability of real-time energy management strategy are validated through the comparison with the stationary control strategy based on initial transition probability matrix (TPM) generated from a long naturalistic driving cycle in the simulation. Results indicate that proposed method has better fuel economy than stationary one and is more effective in real-time control.

Improvements of the Penalty Avoiding Rational Policy Making Algorithm and an Application to the Othello Game

NASA Astrophysics Data System (ADS)

Miyazaki, Kazuteru; Tsuboi, Sougo; Kobayashi, Shigenobu

The purpose of reinforcement learning is to learn an optimal policy in general. However, in 2-players games such as the othello game, it is important to acquire a penalty avoiding policy. In this paper, we focus on formation of a penalty avoiding policy based on the Penalty Avoiding Rational Policy Making algorithm [Miyazaki 01]. In applying it to large-scale problems, we are confronted with the curse of dimensionality. We introduce several ideas and heuristics to overcome the combinational explosion in large-scale problems. First, we propose an algorithm to save the memory by calculation of state transition. Second, we describe how to restrict exploration by two type knowledge; KIFU database and evaluation funcion. We show that our learning player can always defeat against the well-known othello game program KITTY.

Efficiency Improvement of Action Acquisition in Two-Link Robot Arm Using Fuzzy ART with Genetic Algorithm

NASA Astrophysics Data System (ADS)

Kotani, Naoki; Taniguchi, Kenji

An efficient learning method using Fuzzy ART with Genetic Algorithm is proposed. The proposed method reduces the number of trials by using a policy acquired in other tasks because a reinforcement learning needs a lot of the number of trials until an agent acquires appropriate actions. Fuzzy ART is an incremental unsupervised learning algorithm in responce to arbitrary sequences of analog or binary input vectors. Our proposed method gives a policy by crossover or mutation when an agent observes unknown states. Selection controls the category proliferation problem of Fuzzy ART. The effectiveness of the proposed method was verified with the simulation of the reaching problem for the two-link robot arm. The proposed method achieves a reduction of both the number of trials and the number of states.

Quantum machine learning: a classical perspective

NASA Astrophysics Data System (ADS)

Ciliberto, Carlo; Herbster, Mark; Ialongo, Alessandro Davide; Pontil, Massimiliano; Rocchetto, Andrea; Severini, Simone; Wossnig, Leonard

2018-01-01

Recently, increased computational power and data availability, as well as algorithmic advances, have led machine learning (ML) techniques to impressive results in regression, classification, data generation and reinforcement learning tasks. Despite these successes, the proximity to the physical limits of chip fabrication alongside the increasing size of datasets is motivating a growing number of researchers to explore the possibility of harnessing the power of quantum computation to speed up classical ML algorithms. Here we review the literature in quantum ML and discuss perspectives for a mixed readership of classical ML and quantum computation experts. Particular emphasis will be placed on clarifying the limitations of quantum algorithms, how they compare with their best classical counterparts and why quantum resources are expected to provide advantages for learning problems. Learning in the presence of noise and certain computationally hard problems in ML are identified as promising directions for the field. Practical questions, such as how to upload classical data into quantum form, will also be addressed.

Quantum machine learning: a classical perspective

PubMed Central

Ciliberto, Carlo; Herbster, Mark; Ialongo, Alessandro Davide; Pontil, Massimiliano; Severini, Simone; Wossnig, Leonard

2018-01-01

Recently, increased computational power and data availability, as well as algorithmic advances, have led machine learning (ML) techniques to impressive results in regression, classification, data generation and reinforcement learning tasks. Despite these successes, the proximity to the physical limits of chip fabrication alongside the increasing size of datasets is motivating a growing number of researchers to explore the possibility of harnessing the power of quantum computation to speed up classical ML algorithms. Here we review the literature in quantum ML and discuss perspectives for a mixed readership of classical ML and quantum computation experts. Particular emphasis will be placed on clarifying the limitations of quantum algorithms, how they compare with their best classical counterparts and why quantum resources are expected to provide advantages for learning problems. Learning in the presence of noise and certain computationally hard problems in ML are identified as promising directions for the field. Practical questions, such as how to upload classical data into quantum form, will also be addressed. PMID:29434508

Quantum machine learning: a classical perspective.

PubMed

Ciliberto, Carlo; Herbster, Mark; Ialongo, Alessandro Davide; Pontil, Massimiliano; Rocchetto, Andrea; Severini, Simone; Wossnig, Leonard

2018-01-01

Recently, increased computational power and data availability, as well as algorithmic advances, have led machine learning (ML) techniques to impressive results in regression, classification, data generation and reinforcement learning tasks. Despite these successes, the proximity to the physical limits of chip fabrication alongside the increasing size of datasets is motivating a growing number of researchers to explore the possibility of harnessing the power of quantum computation to speed up classical ML algorithms. Here we review the literature in quantum ML and discuss perspectives for a mixed readership of classical ML and quantum computation experts. Particular emphasis will be placed on clarifying the limitations of quantum algorithms, how they compare with their best classical counterparts and why quantum resources are expected to provide advantages for learning problems. Learning in the presence of noise and certain computationally hard problems in ML are identified as promising directions for the field. Practical questions, such as how to upload classical data into quantum form, will also be addressed.

Optimal and Scalable Caching for 5G Using Reinforcement Learning of Space-Time Popularities

NASA Astrophysics Data System (ADS)

Sadeghi, Alireza; Sheikholeslami, Fatemeh; Giannakis, Georgios B.

2018-02-01

Small basestations (SBs) equipped with caching units have potential to handle the unprecedented demand growth in heterogeneous networks. Through low-rate, backhaul connections with the backbone, SBs can prefetch popular files during off-peak traffic hours, and service them to the edge at peak periods. To intelligently prefetch, each SB must learn what and when to cache, while taking into account SB memory limitations, the massive number of available contents, the unknown popularity profiles, as well as the space-time popularity dynamics of user file requests. In this work, local and global Markov processes model user requests, and a reinforcement learning (RL) framework is put forth for finding the optimal caching policy when the transition probabilities involved are unknown. Joint consideration of global and local popularity demands along with cache-refreshing costs allow for a simple, yet practical asynchronous caching approach. The novel RL-based caching relies on a Q-learning algorithm to implement the optimal policy in an online fashion, thus enabling the cache control unit at the SB to learn, track, and possibly adapt to the underlying dynamics. To endow the algorithm with scalability, a linear function approximation of the proposed Q-learning scheme is introduced, offering faster convergence as well as reduced complexity and memory requirements. Numerical tests corroborate the merits of the proposed approach in various realistic settings.

Impedance learning for robotic contact tasks using natural actor-critic algorithm.

PubMed

Kim, Byungchan; Park, Jooyoung; Park, Shinsuk; Kang, Sungchul

2010-04-01

Compared with their robotic counterparts, humans excel at various tasks by using their ability to adaptively modulate arm impedance parameters. This ability allows us to successfully perform contact tasks even in uncertain environments. This paper considers a learning strategy of motor skill for robotic contact tasks based on a human motor control theory and machine learning schemes. Our robot learning method employs impedance control based on the equilibrium point control theory and reinforcement learning to determine the impedance parameters for contact tasks. A recursive least-square filter-based episodic natural actor-critic algorithm is used to find the optimal impedance parameters. The effectiveness of the proposed method was tested through dynamic simulations of various contact tasks. The simulation results demonstrated that the proposed method optimizes the performance of the contact tasks in uncertain conditions of the environment.

Using Reinforcement Learning to Provide Stable Brain-Machine Interface Control Despite Neural Input Reorganization

PubMed Central

Pohlmeyer, Eric A.; Mahmoudi, Babak; Geng, Shijia; Prins, Noeline W.; Sanchez, Justin C.

2014-01-01

Brain-machine interface (BMI) systems give users direct neural control of robotic, communication, or functional electrical stimulation systems. As BMI systems begin transitioning from laboratory settings into activities of daily living, an important goal is to develop neural decoding algorithms that can be calibrated with a minimal burden on the user, provide stable control for long periods of time, and can be responsive to fluctuations in the decoder’s neural input space (e.g. neurons appearing or being lost amongst electrode recordings). These are significant challenges for static neural decoding algorithms that assume stationary input/output relationships. Here we use an actor-critic reinforcement learning architecture to provide an adaptive BMI controller that can successfully adapt to dramatic neural reorganizations, can maintain its performance over long time periods, and which does not require the user to produce specific kinetic or kinematic activities to calibrate the BMI. Two marmoset monkeys used the Reinforcement Learning BMI (RLBMI) to successfully control a robotic arm during a two-target reaching task. The RLBMI was initialized using random initial conditions, and it quickly learned to control the robot from brain states using only a binary evaluative feedback regarding whether previously chosen robot actions were good or bad. The RLBMI was able to maintain control over the system throughout sessions spanning multiple weeks. Furthermore, the RLBMI was able to quickly adapt and maintain control of the robot despite dramatic perturbations to the neural inputs, including a series of tests in which the neuron input space was deliberately halved or doubled. PMID:24498055

Optimal medication dosing from suboptimal clinical examples: a deep reinforcement learning approach.

PubMed

Nemati, Shamim; Ghassemi, Mohammad M; Clifford, Gari D

2016-08-01

Misdosing medications with sensitive therapeutic windows, such as heparin, can place patients at unnecessary risk, increase length of hospital stay, and lead to wasted hospital resources. In this work, we present a clinician-in-the-loop sequential decision making framework, which provides an individualized dosing policy adapted to each patient's evolving clinical phenotype. We employed retrospective data from the publicly available MIMIC II intensive care unit database, and developed a deep reinforcement learning algorithm that learns an optimal heparin dosing policy from sample dosing trails and their associated outcomes in large electronic medical records. Using separate training and testing datasets, our model was observed to be effective in proposing heparin doses that resulted in better expected outcomes than the clinical guidelines. Our results demonstrate that a sequential modeling approach, learned from retrospective data, could potentially be used at the bedside to derive individualized patient dosing policies.

Interconnected growing self-organizing maps for auditory and semantic acquisition modeling

PubMed Central

Cao, Mengxue; Li, Aijun; Fang, Qiang; Kaufmann, Emily; Kröger, Bernd J.

2014-01-01

Based on the incremental nature of knowledge acquisition, in this study we propose a growing self-organizing neural network approach for modeling the acquisition of auditory and semantic categories. We introduce an Interconnected Growing Self-Organizing Maps (I-GSOM) algorithm, which takes associations between auditory information and semantic information into consideration, in this paper. Direct phonetic–semantic association is simulated in order to model the language acquisition in early phases, such as the babbling and imitation stages, in which no phonological representations exist. Based on the I-GSOM algorithm, we conducted experiments using paired acoustic and semantic training data. We use a cyclical reinforcing and reviewing training procedure to model the teaching and learning process between children and their communication partners. A reinforcing-by-link training procedure and a link-forgetting procedure are introduced to model the acquisition of associative relations between auditory and semantic information. Experimental results indicate that (1) I-GSOM has good ability to learn auditory and semantic categories presented within the training data; (2) clear auditory and semantic boundaries can be found in the network representation; (3) cyclical reinforcing and reviewing training leads to a detailed categorization as well as to a detailed clustering, while keeping the clusters that have already been learned and the network structure that has already been developed stable; and (4) reinforcing-by-link training leads to well-perceived auditory–semantic associations. Our I-GSOM model suggests that it is important to associate auditory information with semantic information during language acquisition. Despite its high level of abstraction, our I-GSOM approach can be interpreted as a biologically-inspired neurocomputational model. PMID:24688478

Modeling Humans as Reinforcement Learners: How to Predict Human Behavior in Multi-Stage Games

NASA Technical Reports Server (NTRS)

Lee, Ritchie; Wolpert, David H.; Backhaus, Scott; Bent, Russell; Bono, James; Tracey, Brendan

2011-01-01

This paper introduces a novel framework for modeling interacting humans in a multi-stage game environment by combining concepts from game theory and reinforcement learning. The proposed model has the following desirable characteristics: (1) Bounded rational players, (2) strategic (i.e., players account for one anothers reward functions), and (3) is computationally feasible even on moderately large real-world systems. To do this we extend level-K reasoning to policy space to, for the first time, be able to handle multiple time steps. This allows us to decompose the problem into a series of smaller ones where we can apply standard reinforcement learning algorithms. We investigate these ideas in a cyber-battle scenario over a smart power grid and discuss the relationship between the behavior predicted by our model and what one might expect of real human defenders and attackers.

Functional Contour-following via Haptic Perception and Reinforcement Learning.

PubMed

Hellman, Randall B; Tekin, Cem; van der Schaar, Mihaela; Santos, Veronica J

2018-01-01

Many tasks involve the fine manipulation of objects despite limited visual feedback. In such scenarios, tactile and proprioceptive feedback can be leveraged for task completion. We present an approach for real-time haptic perception and decision-making for a haptics-driven, functional contour-following task: the closure of a ziplock bag. This task is challenging for robots because the bag is deformable, transparent, and visually occluded by artificial fingertip sensors that are also compliant. A deep neural net classifier was trained to estimate the state of a zipper within a robot's pinch grasp. A Contextual Multi-Armed Bandit (C-MAB) reinforcement learning algorithm was implemented to maximize cumulative rewards by balancing exploration versus exploitation of the state-action space. The C-MAB learner outperformed a benchmark Q-learner by more efficiently exploring the state-action space while learning a hard-to-code task. The learned C-MAB policy was tested with novel ziplock bag scenarios and contours (wire, rope). Importantly, this work contributes to the development of reinforcement learning approaches that account for limited resources such as hardware life and researcher time. As robots are used to perform complex, physically interactive tasks in unstructured or unmodeled environments, it becomes important to develop methods that enable efficient and effective learning with physical testbeds.

Intelligent deflection routing in buffer-less networks.

PubMed

Haeri, Soroush; Trajković, Ljiljana

2015-02-01

Deflection routing is employed to ameliorate packet loss caused by contention in buffer-less architectures such as optical burst-switched networks. The main goal of deflection routing is to successfully deflect a packet based only on a limited knowledge that network nodes possess about their environment. In this paper, we present a framework that introduces intelligence to deflection routing (iDef). iDef decouples the design of the signaling infrastructure from the underlying learning algorithm. It consists of a signaling and a decision-making module. Signaling module implements a feedback management protocol while the decision-making module implements a reinforcement learning algorithm. We also propose several learning-based deflection routing protocols, implement them in iDef using the ns-3 network simulator, and compare their performance.

Implementation study of an analog spiking neural network for assisting cardiac delay prediction in a cardiac resynchronization therapy device.

PubMed

Sun, Qing; Schwartz, François; Michel, Jacques; Herve, Yannick; Dalmolin, Renzo

2011-06-01

In this paper, we aim at developing an analog spiking neural network (SNN) for reinforcing the performance of conventional cardiac resynchronization therapy (CRT) devices (also called biventricular pacemakers). Targeting an alternative analog solution in 0.13- μm CMOS technology, this paper proposes an approach to improve cardiac delay predictions in every cardiac period in order to assist the CRT device to provide real-time optimal heartbeats. The primary analog SNN architecture is proposed and its implementation is studied to fulfill the requirement of very low energy consumption. By using the Hebbian learning and reinforcement learning algorithms, the intended adaptive CRT device works with different functional modes. The simulations of both learning algorithms have been carried out, and they were shown to demonstrate the global functionalities. To improve the realism of the system, we introduce various heart behavior models (with constant/variable heart rates) that allow pathologic simulations with/without noise on the signals of the input sensors. The simulations of the global system (pacemaker models coupled with heart models) have been investigated and used to validate the analog spiking neural network implementation.

Integral reinforcement learning for continuous-time input-affine nonlinear systems with simultaneous invariant explorations.

PubMed

Lee, Jae Young; Park, Jin Bae; Choi, Yoon Ho

2015-05-01

This paper focuses on a class of reinforcement learning (RL) algorithms, named integral RL (I-RL), that solve continuous-time (CT) nonlinear optimal control problems with input-affine system dynamics. First, we extend the concepts of exploration, integral temporal difference, and invariant admissibility to the target CT nonlinear system that is governed by a control policy plus a probing signal called an exploration. Then, we show input-to-state stability (ISS) and invariant admissibility of the closed-loop systems with the policies generated by integral policy iteration (I-PI) or invariantly admissible PI (IA-PI) method. Based on these, three online I-RL algorithms named explorized I-PI and integral Q -learning I, II are proposed, all of which generate the same convergent sequences as I-PI and IA-PI under the required excitation condition on the exploration. All the proposed methods are partially or completely model free, and can simultaneously explore the state space in a stable manner during the online learning processes. ISS, invariant admissibility, and convergence properties of the proposed methods are also investigated, and related with these, we show the design principles of the exploration for safe learning. Neural-network-based implementation methods for the proposed schemes are also presented in this paper. Finally, several numerical simulations are carried out to verify the effectiveness of the proposed methods.

Reinforcement learning for partially observable dynamic processes: adaptive dynamic programming using measured output data.

PubMed

Lewis, F L; Vamvoudakis, Kyriakos G

2011-02-01

Approximate dynamic programming (ADP) is a class of reinforcement learning methods that have shown their importance in a variety of applications, including feedback control of dynamical systems. ADP generally requires full information about the system internal states, which is usually not available in practical situations. In this paper, we show how to implement ADP methods using only measured input/output data from the system. Linear dynamical systems with deterministic behavior are considered herein, which are systems of great interest in the control system community. In control system theory, these types of methods are referred to as output feedback (OPFB). The stochastic equivalent of the systems dealt with in this paper is a class of partially observable Markov decision processes. We develop both policy iteration and value iteration algorithms that converge to an optimal controller that requires only OPFB. It is shown that, similar to Q -learning, the new methods have the important advantage that knowledge of the system dynamics is not needed for the implementation of these learning algorithms or for the OPFB control. Only the order of the system, as well as an upper bound on its "observability index," must be known. The learned OPFB controller is in the form of a polynomial autoregressive moving-average controller that has equivalent performance with the optimal state variable feedback gain.

Reinforcement Learning in a Nonstationary Environment: The El Farol Problem

NASA Technical Reports Server (NTRS)

Bell, Ann Maria

1999-01-01

This paper examines the performance of simple learning rules in a complex adaptive system based on a coordination problem modeled on the El Farol problem. The key features of the El Farol problem are that it typically involves a medium number of agents and that agents' pay-off functions have a discontinuous response to increased congestion. First we consider a single adaptive agent facing a stationary environment. We demonstrate that the simple learning rules proposed by Roth and Er'ev can be extremely sensitive to small changes in the initial conditions and that events early in a simulation can affect the performance of the rule over a relatively long time horizon. In contrast, a reinforcement learning rule based on standard practice in the computer science literature converges rapidly and robustly. The situation is reversed when multiple adaptive agents interact: the RE algorithms often converge rapidly to a stable average aggregate attendance despite the slow and erratic behavior of individual learners, while the CS based learners frequently over-attend in the early and intermediate terms. The symmetric mixed strategy equilibria is unstable: all three learning rules ultimately tend towards pure strategies or stabilize in the medium term at non-equilibrium probabilities of attendance. The brittleness of the algorithms in different contexts emphasize the importance of thorough and thoughtful examination of simulation-based results.

Optimizing Chemical Reactions with Deep Reinforcement Learning.

PubMed

Zhou, Zhenpeng; Li, Xiaocheng; Zare, Richard N

2017-12-27

Deep reinforcement learning was employed to optimize chemical reactions. Our model iteratively records the results of a chemical reaction and chooses new experimental conditions to improve the reaction outcome. This model outperformed a state-of-the-art blackbox optimization algorithm by using 71% fewer steps on both simulations and real reactions. Furthermore, we introduced an efficient exploration strategy by drawing the reaction conditions from certain probability distributions, which resulted in an improvement on regret from 0.062 to 0.039 compared with a deterministic policy. Combining the efficient exploration policy with accelerated microdroplet reactions, optimal reaction conditions were determined in 30 min for the four reactions considered, and a better understanding of the factors that control microdroplet reactions was reached. Moreover, our model showed a better performance after training on reactions with similar or even dissimilar underlying mechanisms, which demonstrates its learning ability.

Reinforcement learning for a biped robot based on a CPG-actor-critic method.

PubMed

Nakamura, Yutaka; Mori, Takeshi; Sato, Masa-aki; Ishii, Shin

2007-08-01

Animals' rhythmic movements, such as locomotion, are considered to be controlled by neural circuits called central pattern generators (CPGs), which generate oscillatory signals. Motivated by this biological mechanism, studies have been conducted on the rhythmic movements controlled by CPG. As an autonomous learning framework for a CPG controller, we propose in this article a reinforcement learning method we call the "CPG-actor-critic" method. This method introduces a new architecture to the actor, and its training is roughly based on a stochastic policy gradient algorithm presented recently. We apply this method to an automatic acquisition problem of control for a biped robot. Computer simulations show that training of the CPG can be successfully performed by our method, thus allowing the biped robot to not only walk stably but also adapt to environmental changes.

A Comparative Analysis of Reinforcement Learning Methods

DTIC Science & Technology

1991-10-01

Technology. Support for this research was provided in part by the Mazda Corporation, in part by the University Research Initiative under Office of Naval...results in an update rule (e.g. [Goldberg 89]Goldberg85), genetic algorithms which disregards all history accumulated in the current will not be addressed

Optimizing microstimulation using a reinforcement learning framework.

PubMed

Brockmeier, Austin J; Choi, John S; Distasio, Marcello M; Francis, Joseph T; Príncipe, José C

2011-01-01

The ability to provide sensory feedback is desired to enhance the functionality of neuroprosthetics. Somatosensory feedback provides closed-loop control to the motor system, which is lacking in feedforward neuroprosthetics. In the case of existing somatosensory function, a template of the natural response can be used as a template of desired response elicited by electrical microstimulation. In the case of no initial training data, microstimulation parameters that produce responses close to the template must be selected in an online manner. We propose using reinforcement learning as a framework to balance the exploration of the parameter space and the continued selection of promising parameters for further stimulation. This approach avoids an explicit model of the neural response from stimulation. We explore a preliminary architecture--treating the task as a k-armed bandit--using offline data recorded for natural touch and thalamic microstimulation, and we examine the methods efficiency in exploring the parameter space while concentrating on promising parameter forms. The best matching stimulation parameters, from k = 68 different forms, are selected by the reinforcement learning algorithm consistently after 334 realizations.

Learning tactile skills through curious exploration

PubMed Central

Pape, Leo; Oddo, Calogero M.; Controzzi, Marco; Cipriani, Christian; Förster, Alexander; Carrozza, Maria C.; Schmidhuber, Jürgen

2012-01-01

We present curiosity-driven, autonomous acquisition of tactile exploratory skills on a biomimetic robot finger equipped with an array of microelectromechanical touch sensors. Instead of building tailored algorithms for solving a specific tactile task, we employ a more general curiosity-driven reinforcement learning approach that autonomously learns a set of motor skills in absence of an explicit teacher signal. In this approach, the acquisition of skills is driven by the information content of the sensory input signals relative to a learner that aims at representing sensory inputs using fewer and fewer computational resources. We show that, from initially random exploration of its environment, the robotic system autonomously develops a small set of basic motor skills that lead to different kinds of tactile input. Next, the system learns how to exploit the learned motor skills to solve supervised texture classification tasks. Our approach demonstrates the feasibility of autonomous acquisition of tactile skills on physical robotic platforms through curiosity-driven reinforcement learning, overcomes typical difficulties of engineered solutions for active tactile exploration and underactuated control, and provides a basis for studying developmental learning through intrinsic motivation in robots. PMID:22837748

Tuning fuzzy PD and PI controllers using reinforcement learning.

PubMed

Boubertakh, Hamid; Tadjine, Mohamed; Glorennec, Pierre-Yves; Labiod, Salim

2010-10-01

In this paper, we propose a new auto-tuning fuzzy PD and PI controllers using reinforcement Q-learning (QL) algorithm for SISO (single-input single-output) and TITO (two-input two-output) systems. We first, investigate the design parameters and settings of a typical class of Fuzzy PD (FPD) and Fuzzy PI (FPI) controllers: zero-order Takagi-Sugeno controllers with equidistant triangular membership functions for inputs, equidistant singleton membership functions for output, Larsen's implication method, and average sum defuzzification method. Secondly, the analytical structures of these typical fuzzy PD and PI controllers are compared to their classical counterpart PD and PI controllers. Finally, the effectiveness of the proposed method is proven through simulation examples. Copyright © 2010 ISA. Published by Elsevier Ltd. All rights reserved.

Reinforcement learning interfaces for biomedical database systems.

PubMed

Rudowsky, I; Kulyba, O; Kunin, M; Parsons, S; Raphan, T

2006-01-01

Studies of neural function that are carried out in different laboratories and that address different questions use a wide range of descriptors for data storage, depending on the laboratory and the individuals that input the data. A common approach to describe non-textual data that are referenced through a relational database is to use metadata descriptors. We have recently designed such a prototype system, but to maintain efficiency and a manageable metadata table, free formatted fields were designed as table entries. The database interface application utilizes an intelligent agent to improve integrity of operation. The purpose of this study was to investigate how reinforcement learning algorithms can assist the user in interacting with the database interface application that has been developed to improve the performance of the system.

Improving Robot Locomotion Through Learning Methods for Expensive Black-Box Systems

DTIC Science & Technology

2013-11-01

development of a class of “gradient free” optimization techniques; these include local approaches, such as a Nelder- Mead simplex search (c.f. [73]), and global...1Note that this simple method differs from the Nelder Mead constrained nonlinear optimization method [73]. 39 the Non-dominated Sorting Genetic Algorithm...Kober, and Jan Peters. Model-free inverse reinforcement learning. In International Conference on Artificial Intelligence and Statistics, 2011. [12] George

Learning Sequences of Actions in Collectives of Autonomous Agents

NASA Technical Reports Server (NTRS)

Turner, Kagan; Agogino, Adrian K.; Wolpert, David H.; Clancy, Daniel (Technical Monitor)

2001-01-01

In this paper we focus on the problem of designing a collective of autonomous agents that individually learn sequences of actions such that the resultant sequence of joint actions achieves a predetermined global objective. We are particularly interested in instances of this problem where centralized control is either impossible or impractical. For single agent systems in similar domains, machine learning methods (e.g., reinforcement learners) have been successfully used. However, applying such solutions directly to multi-agent systems often proves problematic, as agents may work at cross-purposes, or have difficulty in evaluating their contribution to achievement of the global objective, or both. Accordingly, the crucial design step in multiagent systems centers on determining the private objectives of each agent so that as the agents strive for those objectives, the system reaches a good global solution. In this work we consider a version of this problem involving multiple autonomous agents in a grid world. We use concepts from collective intelligence to design goals for the agents that are 'aligned' with the global goal, and are 'learnable' in that agents can readily see how their behavior affects their utility. We show that reinforcement learning agents using those goals outperform both 'natural' extensions of single agent algorithms and global reinforcement, learning solutions based on 'team games'.

Optimizing Chemical Reactions with Deep Reinforcement Learning

PubMed Central

2017-01-01

Deep reinforcement learning was employed to optimize chemical reactions. Our model iteratively records the results of a chemical reaction and chooses new experimental conditions to improve the reaction outcome. This model outperformed a state-of-the-art blackbox optimization algorithm by using 71% fewer steps on both simulations and real reactions. Furthermore, we introduced an efficient exploration strategy by drawing the reaction conditions from certain probability distributions, which resulted in an improvement on regret from 0.062 to 0.039 compared with a deterministic policy. Combining the efficient exploration policy with accelerated microdroplet reactions, optimal reaction conditions were determined in 30 min for the four reactions considered, and a better understanding of the factors that control microdroplet reactions was reached. Moreover, our model showed a better performance after training on reactions with similar or even dissimilar underlying mechanisms, which demonstrates its learning ability. PMID:29296675

Extreme Trust Region Policy Optimization for Active Object Recognition.

PubMed

Liu, Huaping; Wu, Yupei; Sun, Fuchun; Huaping Liu; Yupei Wu; Fuchun Sun; Sun, Fuchun; Liu, Huaping; Wu, Yupei

2018-06-01

In this brief, we develop a deep reinforcement learning method to actively recognize objects by choosing a sequence of actions for an active camera that helps to discriminate between the objects. The method is realized using trust region policy optimization, in which the policy is realized by an extreme learning machine and, therefore, leads to efficient optimization algorithm. The experimental results on the publicly available data set show the advantages of the developed extreme trust region optimization method.

Online selective kernel-based temporal difference learning.

PubMed

Chen, Xingguo; Gao, Yang; Wang, Ruili

2013-12-01

In this paper, an online selective kernel-based temporal difference (OSKTD) learning algorithm is proposed to deal with large scale and/or continuous reinforcement learning problems. OSKTD includes two online procedures: online sparsification and parameter updating for the selective kernel-based value function. A new sparsification method (i.e., a kernel distance-based online sparsification method) is proposed based on selective ensemble learning, which is computationally less complex compared with other sparsification methods. With the proposed sparsification method, the sparsified dictionary of samples is constructed online by checking if a sample needs to be added to the sparsified dictionary. In addition, based on local validity, a selective kernel-based value function is proposed to select the best samples from the sample dictionary for the selective kernel-based value function approximator. The parameters of the selective kernel-based value function are iteratively updated by using the temporal difference (TD) learning algorithm combined with the gradient descent technique. The complexity of the online sparsification procedure in the OSKTD algorithm is O(n). In addition, two typical experiments (Maze and Mountain Car) are used to compare with both traditional and up-to-date O(n) algorithms (GTD, GTD2, and TDC using the kernel-based value function), and the results demonstrate the effectiveness of our proposed algorithm. In the Maze problem, OSKTD converges to an optimal policy and converges faster than both traditional and up-to-date algorithms. In the Mountain Car problem, OSKTD converges, requires less computation time compared with other sparsification methods, gets a better local optima than the traditional algorithms, and converges much faster than the up-to-date algorithms. In addition, OSKTD can reach a competitive ultimate optima compared with the up-to-date algorithms.

Somato-dendritic Synaptic Plasticity and Error-backpropagation in Active Dendrites

PubMed Central

Schiess, Mathieu; Urbanczik, Robert; Senn, Walter

2016-01-01

In the last decade dendrites of cortical neurons have been shown to nonlinearly combine synaptic inputs by evoking local dendritic spikes. It has been suggested that these nonlinearities raise the computational power of a single neuron, making it comparable to a 2-layer network of point neurons. But how these nonlinearities can be incorporated into the synaptic plasticity to optimally support learning remains unclear. We present a theoretically derived synaptic plasticity rule for supervised and reinforcement learning that depends on the timing of the presynaptic, the dendritic and the postsynaptic spikes. For supervised learning, the rule can be seen as a biological version of the classical error-backpropagation algorithm applied to the dendritic case. When modulated by a delayed reward signal, the same plasticity is shown to maximize the expected reward in reinforcement learning for various coding scenarios. Our framework makes specific experimental predictions and highlights the unique advantage of active dendrites for implementing powerful synaptic plasticity rules that have access to downstream information via backpropagation of action potentials. PMID:26841235

Scheduled power tracking control of the wind-storage hybrid system based on the reinforcement learning theory

NASA Astrophysics Data System (ADS)

Li, Ze

2017-09-01

In allusion to the intermittency and uncertainty of the wind electricity, energy storage and wind generator are combined into a hybrid system to improve the controllability of the output power. A scheduled power tracking control method is proposed based on the reinforcement learning theory and Q-learning algorithm. In this method, the state space of the environment is formed with two key factors, i.e. the state of charge of the energy storage and the difference value between the actual wind power and scheduled power, the feasible action is the output power of the energy storage, and the corresponding immediate rewarding function is designed to reflect the rationality of the control action. By interacting with the environment and learning from the immediate reward, the optimal control strategy is gradually formed. After that, it could be applied to the scheduled power tracking control of the hybrid system. Finally, the rationality and validity of the method are verified through simulation examples.

Reinforcement Learning for Constrained Energy Trading Games With Incomplete Information.

PubMed

Wang, Huiwei; Huang, Tingwen; Liao, Xiaofeng; Abu-Rub, Haitham; Chen, Guo

2017-10-01

This paper considers the problem of designing adaptive learning algorithms to seek the Nash equilibrium (NE) of the constrained energy trading game among individually strategic players with incomplete information. In this game, each player uses the learning automaton scheme to generate the action probability distribution based on his/her private information for maximizing his own averaged utility. It is shown that if one of admissible mixed-strategies converges to the NE with probability one, then the averaged utility and trading quantity almost surely converge to their expected ones, respectively. For the given discontinuous pricing function, the utility function has already been proved to be upper semicontinuous and payoff secure which guarantee the existence of the mixed-strategy NE. By the strict diagonal concavity of the regularized Lagrange function, the uniqueness of NE is also guaranteed. Finally, an adaptive learning algorithm is provided to generate the strategy probability distribution for seeking the mixed-strategy NE.

Corticostriatal circuit mechanisms of value-based action selection: Implementation of reinforcement learning algorithms and beyond.

PubMed

Morita, Kenji; Jitsev, Jenia; Morrison, Abigail

2016-09-15

Value-based action selection has been suggested to be realized in the corticostriatal local circuits through competition among neural populations. In this article, we review theoretical and experimental studies that have constructed and verified this notion, and provide new perspectives on how the local-circuit selection mechanisms implement reinforcement learning (RL) algorithms and computations beyond them. The striatal neurons are mostly inhibitory, and lateral inhibition among them has been classically proposed to realize "Winner-Take-All (WTA)" selection of the maximum-valued action (i.e., 'max' operation). Although this view has been challenged by the revealed weakness, sparseness, and asymmetry of lateral inhibition, which suggest more complex dynamics, WTA-like competition could still occur on short time scales. Unlike the striatal circuit, the cortical circuit contains recurrent excitation, which may enable retention or temporal integration of information and probabilistic "soft-max" selection. The striatal "max" circuit and the cortical "soft-max" circuit might co-implement an RL algorithm called Q-learning; the cortical circuit might also similarly serve for other algorithms such as SARSA. In these implementations, the cortical circuit presumably sustains activity representing the executed action, which negatively impacts dopamine neurons so that they can calculate reward-prediction-error. Regarding the suggested more complex dynamics of striatal, as well as cortical, circuits on long time scales, which could be viewed as a sequence of short WTA fragments, computational roles remain open: such a sequence might represent (1) sequential state-action-state transitions, constituting replay or simulation of the internal model, (2) a single state/action by the whole trajectory, or (3) probabilistic sampling of state/action. Copyright © 2016. Published by Elsevier B.V.

Autonomous Performance Monitoring System: Monitoring and Self-Tuning (MAST)

NASA Technical Reports Server (NTRS)

Peterson, Chariya; Ziyad, Nigel A.

2000-01-01

Maintaining the long-term performance of software onboard a spacecraft can be a major factor in the cost of operations. In particular, the task of controlling and maintaining a future mission of distributed spacecraft will undoubtedly pose a great challenge, since the complexity of multiple spacecraft flying in formation grows rapidly as the number of spacecraft in the formation increases. Eventually, new approaches will be required in developing viable control systems that can handle the complexity of the data and that are flexible, reliable and efficient. In this paper we propose a methodology that aims to maintain the accuracy of flight software, while reducing the computational complexity of software tuning tasks. The proposed Monitoring and Self-Tuning (MAST) method consists of two parts: a flight software monitoring algorithm and a tuning algorithm. The dependency on the software being monitored is mostly contained in the monitoring process, while the tuning process is a generic algorithm independent of the detailed knowledge on the software. This architecture will enable MAST to be applicable to different onboard software controlling various dynamics of the spacecraft, such as attitude self-calibration, and formation control. An advantage of MAST over conventional techniques such as filter or batch least square is that the tuning algorithm uses machine learning approach to handle uncertainty in the problem domain, resulting in reducing over all computational complexity. The underlying concept of this technique is a reinforcement learning scheme based on cumulative probability generated by the historical performance of the system. The success of MAST will depend heavily on the reinforcement scheme used in the tuning algorithm, which guarantees the tuning solutions exist.

Reinforcement Learning Models and Their Neural Correlates: An Activation Likelihood Estimation Meta-Analysis

PubMed Central

Kumar, Poornima; Eickhoff, Simon B.; Dombrovski, Alexandre Y.

2015-01-01

Reinforcement learning describes motivated behavior in terms of two abstract signals. The representation of discrepancies between expected and actual rewards/punishments – prediction error – is thought to update the expected value of actions and predictive stimuli. Electrophysiological and lesion studies suggest that mesostriatal prediction error signals control behavior through synaptic modification of cortico-striato-thalamic networks. Signals in the ventromedial prefrontal and orbitofrontal cortex are implicated in representing expected value. To obtain unbiased maps of these representations in the human brain, we performed a meta-analysis of functional magnetic resonance imaging studies that employed algorithmic reinforcement learning models, across a variety of experimental paradigms. We found that the ventral striatum (medial and lateral) and midbrain/thalamus represented reward prediction errors, consistent with animal studies. Prediction error signals were also seen in the frontal operculum/insula, particularly for social rewards. In Pavlovian studies, striatal prediction error signals extended into the amygdala, while instrumental tasks engaged the caudate. Prediction error maps were sensitive to the model-fitting procedure (fixed or individually-estimated) and to the extent of spatial smoothing. A correlate of expected value was found in a posterior region of the ventromedial prefrontal cortex, caudal and medial to the orbitofrontal regions identified in animal studies. These findings highlight a reproducible motif of reinforcement learning in the cortico-striatal loops and identify methodological dimensions that may influence the reproducibility of activation patterns across studies. PMID:25665667

Incorporation of perception-based information in robot learning using fuzzy reinforcement learning agents

NASA Astrophysics Data System (ADS)

Zhou, Changjiu; Meng, Qingchun; Guo, Zhongwen; Qu, Wiefen; Yin, Bo

2002-04-01

Robot learning in unstructured environments has been proved to be an extremely challenging problem, mainly because of many uncertainties always present in the real world. Human beings, on the other hand, seem to cope very well with uncertain and unpredictable environments, often relying on perception-based information. Furthermore, humans beings can also utilize perceptions to guide their learning on those parts of the perception-action space that are actually relevant to the task. Therefore, we conduct a research aimed at improving robot learning through the incorporation of both perception-based and measurement-based information. For this reason, a fuzzy reinforcement learning (FRL) agent is proposed in this paper. Based on a neural-fuzzy architecture, different kinds of information can be incorporated into the FRL agent to initialise its action network, critic network and evaluation feedback module so as to accelerate its learning. By making use of the global optimisation capability of GAs (genetic algorithms), a GA-based FRL (GAFRL) agent is presented to solve the local minima problem in traditional actor-critic reinforcement learning. On the other hand, with the prediction capability of the critic network, GAs can perform a more effective global search. Different GAFRL agents are constructed and verified by using the simulation model of a physical biped robot. The simulation analysis shows that the biped learning rate for dynamic balance can be improved by incorporating perception-based information on biped balancing and walking evaluation. The biped robot can find its application in ocean exploration, detection or sea rescue activity, as well as military maritime activity.

Application of fuzzy logic-neural network based reinforcement learning to proximity and docking operations: Special approach/docking testcase results

NASA Technical Reports Server (NTRS)

Jani, Yashvant

1993-01-01

As part of the RICIS project, the reinforcement learning techniques developed at Ames Research Center are being applied to proximity and docking operations using the Shuttle and Solar Maximum Mission (SMM) satellite simulation. In utilizing these fuzzy learning techniques, we use the Approximate Reasoning based Intelligent Control (ARIC) architecture, and so we use these two terms interchangeably to imply the same. This activity is carried out in the Software Technology Laboratory utilizing the Orbital Operations Simulator (OOS) and programming/testing support from other contractor personnel. This report is the final deliverable D4 in our milestones and project activity. It provides the test results for the special testcase of approach/docking scenario for the shuttle and SMM satellite. Based on our experience and analysis with the attitude and translational controllers, we have modified the basic configuration of the reinforcement learning algorithm in ARIC. The shuttle translational controller and its implementation in ARIC is described in our deliverable D3. In order to simulate the final approach and docking operations, we have set-up this special testcase as described in section 2. The ARIC performance results for these operations are discussed in section 3 and conclusions are provided in section 4 along with the summary for the project.

Learning to soar in turbulent environments

NASA Astrophysics Data System (ADS)

Reddy, Gautam; Celani, Antonio; Sejnowski, Terrence; Vergassola, Massimo

Birds and gliders exploit warm, rising atmospheric currents (thermals) to reach heights comparable to low-lying clouds with a reduced expenditure of energy. Soaring provides a remarkable instance of complex decision-making in biology and requires a long-term strategy to effectively use the ascending thermals. Furthermore, the problem is technologically relevant to extend the flying range of autonomous gliders. The formation of thermals unavoidably generates strong turbulent fluctuations, which make deriving an efficient policy harder and thus constitute an essential element of soaring. Here, we approach soaring flight as a problem of learning to navigate highly fluctuating turbulent environments. We simulate the atmospheric boundary layer by numerical models of turbulent convective flow and combine them with model-free, experience-based, reinforcement learning algorithms to train the virtual gliders. For the learned policies in the regimes of moderate and strong turbulence levels, the virtual glider adopts an increasingly conservative policy as turbulence levels increase, quantifying the degree of risk affordable in turbulent environments. Reinforcement learning uncovers those sensorimotor cues that permit effective control over soaring in turbulent environments.

Learning from Noisy and Delayed Rewards: The Value of Reinforcement Learning to Defense Modeling and Simulation

DTIC Science & Technology

2012-09-01

following 500 trials with 1000 replications with single reward upon attainment of the goal state by algorithm and policy. DQ- C with -greedy obtained...aspects of the civilian population rather than combat forces. These agents rep- resent not a single human, but a population segment. Similar...TD(λ) combines elements of MC and TD methods into a single framework to estimate the value of each state, V(s), through the use of eligibility traces

Segmentation of neuronal structures using SARSA (λ)-based boundary amendment with reinforced gradient-descent curve shape fitting.

PubMed

Zhu, Fei; Liu, Quan; Fu, Yuchen; Shen, Bairong

2014-01-01

The segmentation of structures in electron microscopy (EM) images is very important for neurobiological research. The low resolution neuronal EM images contain noise and generally few features are available for segmentation, therefore application of the conventional approaches to identify the neuron structure from EM images is not successful. We therefore present a multi-scale fused structure boundary detection algorithm in this study. In the algorithm, we generate an EM image Gaussian pyramid first, then at each level of the pyramid, we utilize Laplacian of Gaussian function (LoG) to attain structure boundary, we finally assemble the detected boundaries by using fusion algorithm to attain a combined neuron structure image. Since the obtained neuron structures usually have gaps, we put forward a reinforcement learning-based boundary amendment method to connect the gaps in the detected boundaries. We use a SARSA (λ)-based curve traveling and amendment approach derived from reinforcement learning to repair the incomplete curves. Using this algorithm, a moving point starts from one end of the incomplete curve and walks through the image where the decisions are supervised by the approximated curve model, with the aim of minimizing the connection cost until the gap is closed. Our approach provided stable and efficient structure segmentation. The test results using 30 EM images from ISBI 2012 indicated that both of our approaches, i.e., with or without boundary amendment, performed better than six conventional boundary detection approaches. In particular, after amendment, the Rand error and warping error, which are the most important performance measurements during structure segmentation, were reduced to very low values. The comparison with the benchmark method of ISBI 2012 and the recent developed methods also indicates that our method performs better for the accurate identification of substructures in EM images and therefore useful for the identification of imaging features related to brain diseases.

Segmentation of Neuronal Structures Using SARSA (λ)-Based Boundary Amendment with Reinforced Gradient-Descent Curve Shape Fitting

PubMed Central

Zhu, Fei; Liu, Quan; Fu, Yuchen; Shen, Bairong

2014-01-01

The segmentation of structures in electron microscopy (EM) images is very important for neurobiological research. The low resolution neuronal EM images contain noise and generally few features are available for segmentation, therefore application of the conventional approaches to identify the neuron structure from EM images is not successful. We therefore present a multi-scale fused structure boundary detection algorithm in this study. In the algorithm, we generate an EM image Gaussian pyramid first, then at each level of the pyramid, we utilize Laplacian of Gaussian function (LoG) to attain structure boundary, we finally assemble the detected boundaries by using fusion algorithm to attain a combined neuron structure image. Since the obtained neuron structures usually have gaps, we put forward a reinforcement learning-based boundary amendment method to connect the gaps in the detected boundaries. We use a SARSA (λ)-based curve traveling and amendment approach derived from reinforcement learning to repair the incomplete curves. Using this algorithm, a moving point starts from one end of the incomplete curve and walks through the image where the decisions are supervised by the approximated curve model, with the aim of minimizing the connection cost until the gap is closed. Our approach provided stable and efficient structure segmentation. The test results using 30 EM images from ISBI 2012 indicated that both of our approaches, i.e., with or without boundary amendment, performed better than six conventional boundary detection approaches. In particular, after amendment, the Rand error and warping error, which are the most important performance measurements during structure segmentation, were reduced to very low values. The comparison with the benchmark method of ISBI 2012 and the recent developed methods also indicates that our method performs better for the accurate identification of substructures in EM images and therefore useful for the identification of imaging features related to brain diseases. PMID:24625699

The power of associative learning and the ontogeny of optimal behaviour.

PubMed

Enquist, Magnus; Lind, Johan; Ghirlanda, Stefano

2016-11-01

Behaving efficiently (optimally or near-optimally) is central to animals' adaptation to their environment. Much evolutionary biology assumes, implicitly or explicitly, that optimal behavioural strategies are genetically inherited, yet the behaviour of many animals depends crucially on learning. The question of how learning contributes to optimal behaviour is largely open. Here we propose an associative learning model that can learn optimal behaviour in a wide variety of ecologically relevant circumstances. The model learns through chaining, a term introduced by Skinner to indicate learning of behaviour sequences by linking together shorter sequences or single behaviours. Our model formalizes the concept of conditioned reinforcement (the learning process that underlies chaining) and is closely related to optimization algorithms from machine learning. Our analysis dispels the common belief that associative learning is too limited to produce 'intelligent' behaviour such as tool use, social learning, self-control or expectations of the future. Furthermore, the model readily accounts for both instinctual and learned aspects of behaviour, clarifying how genetic evolution and individual learning complement each other, and bridging a long-standing divide between ethology and psychology. We conclude that associative learning, supported by genetic predispositions and including the oft-neglected phenomenon of conditioned reinforcement, may suffice to explain the ontogeny of optimal behaviour in most, if not all, non-human animals. Our results establish associative learning as a more powerful optimizing mechanism than acknowledged by current opinion.

The power of associative learning and the ontogeny of optimal behaviour

PubMed Central

Enquist, Magnus; Lind, Johan

2016-01-01

Behaving efficiently (optimally or near-optimally) is central to animals' adaptation to their environment. Much evolutionary biology assumes, implicitly or explicitly, that optimal behavioural strategies are genetically inherited, yet the behaviour of many animals depends crucially on learning. The question of how learning contributes to optimal behaviour is largely open. Here we propose an associative learning model that can learn optimal behaviour in a wide variety of ecologically relevant circumstances. The model learns through chaining, a term introduced by Skinner to indicate learning of behaviour sequences by linking together shorter sequences or single behaviours. Our model formalizes the concept of conditioned reinforcement (the learning process that underlies chaining) and is closely related to optimization algorithms from machine learning. Our analysis dispels the common belief that associative learning is too limited to produce ‘intelligent’ behaviour such as tool use, social learning, self-control or expectations of the future. Furthermore, the model readily accounts for both instinctual and learned aspects of behaviour, clarifying how genetic evolution and individual learning complement each other, and bridging a long-standing divide between ethology and psychology. We conclude that associative learning, supported by genetic predispositions and including the oft-neglected phenomenon of conditioned reinforcement, may suffice to explain the ontogeny of optimal behaviour in most, if not all, non-human animals. Our results establish associative learning as a more powerful optimizing mechanism than acknowledged by current opinion. PMID:28018662

Clipping in neurocontrol by adaptive dynamic programming.

PubMed

Fairbank, Michael; Prokhorov, Danil; Alonso, Eduardo

2014-10-01

In adaptive dynamic programming, neurocontrol, and reinforcement learning, the objective is for an agent to learn to choose actions so as to minimize a total cost function. In this paper, we show that when discretized time is used to model the motion of the agent, it can be very important to do clipping on the motion of the agent in the final time step of the trajectory. By clipping, we mean that the final time step of the trajectory is to be truncated such that the agent stops exactly at the first terminal state reached, and no distance further. We demonstrate that when clipping is omitted, learning performance can fail to reach the optimum, and when clipping is done properly, learning performance can improve significantly. The clipping problem we describe affects algorithms that use explicit derivatives of the model functions of the environment to calculate a learning gradient. These include backpropagation through time for control and methods based on dual heuristic programming. However, the clipping problem does not significantly affect methods based on heuristic dynamic programming, temporal differences learning, or policy-gradient learning algorithms.

Beam-column joint shear prediction using hybridized deep learning neural network with genetic algorithm

NASA Astrophysics Data System (ADS)

Mundher Yaseen, Zaher; Abdulmohsin Afan, Haitham; Tran, Minh-Tung

2018-04-01

Scientifically evidenced that beam-column joints are a critical point in the reinforced concrete (RC) structure under the fluctuation loads effects. In this novel hybrid data-intelligence model developed to predict the joint shear behavior of exterior beam-column structure frame. The hybrid data-intelligence model is called genetic algorithm integrated with deep learning neural network model (GA-DLNN). The genetic algorithm is used as prior modelling phase for the input approximation whereas the DLNN predictive model is used for the prediction phase. To demonstrate this structural problem, experimental data is collected from the literature that defined the dimensional and specimens’ properties. The attained findings evidenced the efficitveness of the hybrid GA-DLNN in modelling beam-column joint shear problem. In addition, the accurate prediction achived with less input variables owing to the feasibility of the evolutionary phase.

Tracked robot controllers for climbing obstacles autonomously

NASA Astrophysics Data System (ADS)

Vincent, Isabelle

2009-05-01

Research in mobile robot navigation has demonstrated some success in navigating flat indoor environments while avoiding obstacles. However, the challenge of analyzing complex environments to climb obstacles autonomously has had very little success due to the complexity of the task. Unmanned ground vehicles currently exhibit simple autonomous behaviours compared to the human ability to move in the world. This paper presents the control algorithms designed for a tracked mobile robot to autonomously climb obstacles by varying its tracks configuration. Two control algorithms are proposed to solve the autonomous locomotion problem for climbing obstacles. First, a reactive controller evaluates the appropriate geometric configuration based on terrain and vehicle geometric considerations. Then, a reinforcement learning algorithm finds alternative solutions when the reactive controller gets stuck while climbing an obstacle. The methodology combines reactivity to learning. The controllers have been demonstrated in box and stair climbing simulations. The experiments illustrate the effectiveness of the proposed approach for crossing obstacles.

An analysis of value function learning with piecewise linear control

NASA Astrophysics Data System (ADS)

Tutsoy, Onder; Brown, Martin

2016-05-01

Reinforcement learning (RL) algorithms attempt to learn optimal control actions by iteratively estimating a long-term measure of system performance, the so-called value function. For example, RL algorithms have been applied to walking robots to examine the connection between robot motion and the brain, which is known as embodied cognition. In this paper, RL algorithms are analysed using an exemplar test problem. A closed form solution for the value function is calculated and this is represented in terms of a set of basis functions and parameters, which is used to investigate parameter convergence. The value function expression is shown to have a polynomial form where the polynomial terms depend on the plant's parameters and the value function's discount factor. It is shown that the temporal difference error introduces a null space for the differenced higher order basis associated with the effects of controller switching (saturated to linear control or terminating an experiment) apart from the time of the switch. This leads to slow convergence in the relevant subspace. It is also shown that badly conditioned learning problems can occur, and this is a function of the value function discount factor and the controller switching points. Finally, a comparison is performed between the residual gradient and TD(0) learning algorithms, and it is shown that the former has a faster rate of convergence for this test problem.

Using a Search Engine-Based Mutually Reinforcing Approach to Assess the Semantic Relatedness of Biomedical Terms

PubMed Central

Hsu, Yi-Yu; Chen, Hung-Yu; Kao, Hung-Yu

2013-01-01

Background Determining the semantic relatedness of two biomedical terms is an important task for many text-mining applications in the biomedical field. Previous studies, such as those using ontology-based and corpus-based approaches, measured semantic relatedness by using information from the structure of biomedical literature, but these methods are limited by the small size of training resources. To increase the size of training datasets, the outputs of search engines have been used extensively to analyze the lexical patterns of biomedical terms. Methodology/Principal Findings In this work, we propose the Mutually Reinforcing Lexical Pattern Ranking (ReLPR) algorithm for learning and exploring the lexical patterns of synonym pairs in biomedical text. ReLPR employs lexical patterns and their pattern containers to assess the semantic relatedness of biomedical terms. By combining sentence structures and the linking activities between containers and lexical patterns, our algorithm can explore the correlation between two biomedical terms. Conclusions/Significance The average correlation coefficient of the ReLPR algorithm was 0.82 for various datasets. The results of the ReLPR algorithm were significantly superior to those of previous methods. PMID:24348899

Using Collective Intelligence to Route Internet Traffic

NASA Technical Reports Server (NTRS)

Wolpert, David H.; Tumer, Kagan; Frank, Jeremy

1998-01-01

A Collective Intelligence (COIN) is a community of interacting reinforcement learning (RL) algorithms designed so that their collective behavior maximizes a global utility function. We introduce the theory of COINs, then present experiments using that theory to design COINs to control internet traffic routing. These experiments indicate that COINs outperform previous RL-based systems for such routing that have previously been investigated.

A Novel Dynamic Spectrum Access Framework Based on Reinforcement Learning for Cognitive Radio Sensor Networks.

PubMed

Lin, Yun; Wang, Chao; Wang, Jiaxing; Dou, Zheng

2016-10-12

Cognitive radio sensor networks are one of the kinds of application where cognitive techniques can be adopted and have many potential applications, challenges and future research trends. According to the research surveys, dynamic spectrum access is an important and necessary technology for future cognitive sensor networks. Traditional methods of dynamic spectrum access are based on spectrum holes and they have some drawbacks, such as low accessibility and high interruptibility, which negatively affect the transmission performance of the sensor networks. To address this problem, in this paper a new initialization mechanism is proposed to establish a communication link and set up a sensor network without adopting spectrum holes to convey control information. Specifically, firstly a transmission channel model for analyzing the maximum accessible capacity for three different polices in a fading environment is discussed. Secondly, a hybrid spectrum access algorithm based on a reinforcement learning model is proposed for the power allocation problem of both the transmission channel and the control channel. Finally, extensive simulations have been conducted and simulation results show that this new algorithm provides a significant improvement in terms of the tradeoff between the control channel reliability and the efficiency of the transmission channel.

A Novel Dynamic Spectrum Access Framework Based on Reinforcement Learning for Cognitive Radio Sensor Networks

PubMed Central

Lin, Yun; Wang, Chao; Wang, Jiaxing; Dou, Zheng

2016-01-01

Cognitive radio sensor networks are one of the kinds of application where cognitive techniques can be adopted and have many potential applications, challenges and future research trends. According to the research surveys, dynamic spectrum access is an important and necessary technology for future cognitive sensor networks. Traditional methods of dynamic spectrum access are based on spectrum holes and they have some drawbacks, such as low accessibility and high interruptibility, which negatively affect the transmission performance of the sensor networks. To address this problem, in this paper a new initialization mechanism is proposed to establish a communication link and set up a sensor network without adopting spectrum holes to convey control information. Specifically, firstly a transmission channel model for analyzing the maximum accessible capacity for three different polices in a fading environment is discussed. Secondly, a hybrid spectrum access algorithm based on a reinforcement learning model is proposed for the power allocation problem of both the transmission channel and the control channel. Finally, extensive simulations have been conducted and simulation results show that this new algorithm provides a significant improvement in terms of the tradeoff between the control channel reliability and the efficiency of the transmission channel. PMID:27754316

Maximization of Learning Speed Due to Neuronal Redundancy in Reinforcement Learning

NASA Astrophysics Data System (ADS)

Takiyama, Ken

2016-11-01

Adaptable neural activity contributes to the flexibility of human behavior, which is optimized in situations such as motor learning and decision making. Although learning signals in motor learning and decision making are low-dimensional, neural activity, which is very high dimensional, must be modified to achieve optimal performance based on the low-dimensional signal, resulting in a severe credit-assignment problem. Despite this problem, the human brain contains a vast number of neurons, leaving an open question: what is the functional significance of the huge number of neurons? Here, I address this question by analyzing a redundant neural network with a reinforcement-learning algorithm in which the numbers of neurons and output units are N and M, respectively. Because many combinations of neural activity can generate the same output under the condition of N ≫ M, I refer to the index N - M as neuronal redundancy. Although greater neuronal redundancy makes the credit-assignment problem more severe, I demonstrate that a greater degree of neuronal redundancy facilitates learning speed. Thus, in an apparent contradiction of the credit-assignment problem, I propose the hypothesis that a functional role of a huge number of neurons or a huge degree of neuronal redundancy is to facilitate learning speed.

A clustering-based graph Laplacian framework for value function approximation in reinforcement learning.

PubMed

Xu, Xin; Huang, Zhenhua; Graves, Daniel; Pedrycz, Witold

2014-12-01

In order to deal with the sequential decision problems with large or continuous state spaces, feature representation and function approximation have been a major research topic in reinforcement learning (RL). In this paper, a clustering-based graph Laplacian framework is presented for feature representation and value function approximation (VFA) in RL. By making use of clustering-based techniques, that is, K-means clustering or fuzzy C-means clustering, a graph Laplacian is constructed by subsampling in Markov decision processes (MDPs) with continuous state spaces. The basis functions for VFA can be automatically generated from spectral analysis of the graph Laplacian. The clustering-based graph Laplacian is integrated with a class of approximation policy iteration algorithms called representation policy iteration (RPI) for RL in MDPs with continuous state spaces. Simulation and experimental results show that, compared with previous RPI methods, the proposed approach needs fewer sample points to compute an efficient set of basis functions and the learning control performance can be improved for a variety of parameter settings.

Adaptive critic autopilot design of bank-to-turn missiles using fuzzy basis function networks.

PubMed

Lin, Chuan-Kai

2005-04-01

A new adaptive critic autopilot design for bank-to-turn missiles is presented. In this paper, the architecture of adaptive critic learning scheme contains a fuzzy-basis-function-network based associative search element (ASE), which is employed to approximate nonlinear and complex functions of bank-to-turn missiles, and an adaptive critic element (ACE) generating the reinforcement signal to tune the associative search element. In the design of the adaptive critic autopilot, the control law receives signals from a fixed gain controller, an ASE and an adaptive robust element, which can eliminate approximation errors and disturbances. Traditional adaptive critic reinforcement learning is the problem faced by an agent that must learn behavior through trial-and-error interactions with a dynamic environment, however, the proposed tuning algorithm can significantly shorten the learning time by online tuning all parameters of fuzzy basis functions and weights of ASE and ACE. Moreover, the weight updating law derived from the Lyapunov stability theory is capable of guaranteeing both tracking performance and stability. Computer simulation results confirm the effectiveness of the proposed adaptive critic autopilot.

Reinforcement learning design-based adaptive tracking control with less learning parameters for nonlinear discrete-time MIMO systems.

PubMed

Liu, Yan-Jun; Tang, Li; Tong, Shaocheng; Chen, C L Philip; Li, Dong-Juan

2015-01-01

Based on the neural network (NN) approximator, an online reinforcement learning algorithm is proposed for a class of affine multiple input and multiple output (MIMO) nonlinear discrete-time systems with unknown functions and disturbances. In the design procedure, two networks are provided where one is an action network to generate an optimal control signal and the other is a critic network to approximate the cost function. An optimal control signal and adaptation laws can be generated based on two NNs. In the previous approaches, the weights of critic and action networks are updated based on the gradient descent rule and the estimations of optimal weight vectors are directly adjusted in the design. Consequently, compared with the existing results, the main contributions of this paper are: 1) only two parameters are needed to be adjusted, and thus the number of the adaptation laws is smaller than the previous results and 2) the updating parameters do not depend on the number of the subsystems for MIMO systems and the tuning rules are replaced by adjusting the norms on optimal weight vectors in both action and critic networks. It is proven that the tracking errors, the adaptation laws, and the control inputs are uniformly bounded using Lyapunov analysis method. The simulation examples are employed to illustrate the effectiveness of the proposed algorithm.

Reinforce: An Ensemble Approach for Inferring PPI Network from AP-MS Data.

PubMed

Tian, Bo; Duan, Qiong; Zhao, Can; Teng, Ben; He, Zengyou

2017-05-17

Affinity Purification-Mass Spectrometry (AP-MS) is one of the most important technologies for constructing protein-protein interaction (PPI) networks. In this paper, we propose an ensemble method, Reinforce, for inferring PPI network from AP-MS data set. The new algorithm named Reinforce is based on rank aggregation and false discovery rate control. Under the null hypothesis that the interaction scores from different scoring methods are randomly generated, Reinforce follows three steps to integrate multiple ranking results from different algorithms or different data sets. The experimental results show that Reinforce can get more stable and accurate inference results than existing algorithms. The source codes of Reinforce and data sets used in the experiments are available at: https://sourceforge.net/projects/reinforce/.

Learning to soar in turbulent environments

PubMed Central

Reddy, Gautam; Celani, Antonio; Sejnowski, Terrence J.; Vergassola, Massimo

2016-01-01

Birds and gliders exploit warm, rising atmospheric currents (thermals) to reach heights comparable to low-lying clouds with a reduced expenditure of energy. This strategy of flight (thermal soaring) is frequently used by migratory birds. Soaring provides a remarkable instance of complex decision making in biology and requires a long-term strategy to effectively use the ascending thermals. Furthermore, the problem is technologically relevant to extend the flying range of autonomous gliders. Thermal soaring is commonly observed in the atmospheric convective boundary layer on warm, sunny days. The formation of thermals unavoidably generates strong turbulent fluctuations, which constitute an essential element of soaring. Here, we approach soaring flight as a problem of learning to navigate complex, highly fluctuating turbulent environments. We simulate the atmospheric boundary layer by numerical models of turbulent convective flow and combine them with model-free, experience-based, reinforcement learning algorithms to train the gliders. For the learned policies in the regimes of moderate and strong turbulence levels, the glider adopts an increasingly conservative policy as turbulence levels increase, quantifying the degree of risk affordable in turbulent environments. Reinforcement learning uncovers those sensorimotor cues that permit effective control over soaring in turbulent environments. PMID:27482099

Learning to soar in turbulent environments.

PubMed

Reddy, Gautam; Celani, Antonio; Sejnowski, Terrence J; Vergassola, Massimo

2016-08-16

Birds and gliders exploit warm, rising atmospheric currents (thermals) to reach heights comparable to low-lying clouds with a reduced expenditure of energy. This strategy of flight (thermal soaring) is frequently used by migratory birds. Soaring provides a remarkable instance of complex decision making in biology and requires a long-term strategy to effectively use the ascending thermals. Furthermore, the problem is technologically relevant to extend the flying range of autonomous gliders. Thermal soaring is commonly observed in the atmospheric convective boundary layer on warm, sunny days. The formation of thermals unavoidably generates strong turbulent fluctuations, which constitute an essential element of soaring. Here, we approach soaring flight as a problem of learning to navigate complex, highly fluctuating turbulent environments. We simulate the atmospheric boundary layer by numerical models of turbulent convective flow and combine them with model-free, experience-based, reinforcement learning algorithms to train the gliders. For the learned policies in the regimes of moderate and strong turbulence levels, the glider adopts an increasingly conservative policy as turbulence levels increase, quantifying the degree of risk affordable in turbulent environments. Reinforcement learning uncovers those sensorimotor cues that permit effective control over soaring in turbulent environments.

Effective reinforcement learning following cerebellar damage requires a balance between exploration and motor noise.

PubMed

Therrien, Amanda S; Wolpert, Daniel M; Bastian, Amy J

2016-01-01

Reinforcement and error-based processes are essential for motor learning, with the cerebellum thought to be required only for the error-based mechanism. Here we examined learning and retention of a reaching skill under both processes. Control subjects learned similarly from reinforcement and error-based feedback, but showed much better retention under reinforcement. To apply reinforcement to cerebellar patients, we developed a closed-loop reinforcement schedule in which task difficulty was controlled based on recent performance. This schedule produced substantial learning in cerebellar patients and controls. Cerebellar patients varied in their learning under reinforcement but fully retained what was learned. In contrast, they showed complete lack of retention in error-based learning. We developed a mechanistic model of the reinforcement task and found that learning depended on a balance between exploration variability and motor noise. While the cerebellar and control groups had similar exploration variability, the patients had greater motor noise and hence learned less. Our results suggest that cerebellar damage indirectly impairs reinforcement learning by increasing motor noise, but does not interfere with the reinforcement mechanism itself. Therefore, reinforcement can be used to learn and retain novel skills, but optimal reinforcement learning requires a balance between exploration variability and motor noise. © The Author (2015). Published by Oxford University Press on behalf of the Guarantors of Brain.

Effective reinforcement learning following cerebellar damage requires a balance between exploration and motor noise

PubMed Central

Therrien, Amanda S.; Wolpert, Daniel M.

2016-01-01

Abstract See Miall and Galea (doi: 10.1093/awv343 ) for a scientific commentary on this article. Reinforcement and error-based processes are essential for motor learning, with the cerebellum thought to be required only for the error-based mechanism. Here we examined learning and retention of a reaching skill under both processes. Control subjects learned similarly from reinforcement and error-based feedback, but showed much better retention under reinforcement. To apply reinforcement to cerebellar patients, we developed a closed-loop reinforcement schedule in which task difficulty was controlled based on recent performance. This schedule produced substantial learning in cerebellar patients and controls. Cerebellar patients varied in their learning under reinforcement but fully retained what was learned. In contrast, they showed complete lack of retention in error-based learning. We developed a mechanistic model of the reinforcement task and found that learning depended on a balance between exploration variability and motor noise. While the cerebellar and control groups had similar exploration variability, the patients had greater motor noise and hence learned less. Our results suggest that cerebellar damage indirectly impairs reinforcement learning by increasing motor noise, but does not interfere with the reinforcement mechanism itself. Therefore, reinforcement can be used to learn and retain novel skills, but optimal reinforcement learning requires a balance between exploration variability and motor noise. PMID:26626368

Apprenticeship Learning: Learning to Schedule from Human Experts

DTIC Science & Technology

2016-06-09

approaches to learning such models are based on Markov models, such as reinforcement learning or inverse reinforcement learning (Busoniu, Babuska, and De...via inverse reinforcement learning. In ICML. Barto, A. G., and Mahadevan, S. 2003. Recent advances in hierarchical reinforcement learning. Discrete...of tasks with temporal constraints. In Proc. AAAI, 2110–2116. Odom, P., and Natarajan, S. 2015. Active advice seeking for inverse reinforcement

Hierarchical Traces for Reduced NSM Memory Requirements

NASA Astrophysics Data System (ADS)

Dahl, Torbjørn S.

This paper presents work on using hierarchical long term memory to reduce the memory requirements of nearest sequence memory (NSM) learning, a previously published, instance-based reinforcement learning algorithm. A hierarchical memory representation reduces the memory requirements by allowing traces to share common sub-sequences. We present moderated mechanisms for estimating discounted future rewards and for dealing with hidden state using hierarchical memory. We also present an experimental analysis of how the sub-sequence length affects the memory compression achieved and show that the reduced memory requirements do not effect the speed of learning. Finally, we analyse and discuss the persistence of the sub-sequences independent of specific trace instances.

Wireless, amphibious theory for reinforcement learning

NASA Astrophysics Data System (ADS)

Li, Jinci

2013-10-01

Cryptographers agree that heterogeneous information are an interesting new topic in the field of cryptography, and biologists concur. Given the current status of stochastic epistemologies, security experts clearly desire the construction of flip-flop gates [1, 2, 3]. Mungo, our new system for authenticated algorithms, is the solution to all of these challenges. Though such a hypothesis at first glance seems perverse, it has ample historical precedence.

Cross-language opinion lexicon extraction using mutual-reinforcement label propagation.

PubMed

Lin, Zheng; Tan, Songbo; Liu, Yue; Cheng, Xueqi; Xu, Xueke

2013-01-01

There is a growing interest in automatically building opinion lexicon from sources such as product reviews. Most of these methods depend on abundant external resources such as WordNet, which limits the applicability of these methods. Unsupervised or semi-supervised learning provides an optional solution to multilingual opinion lexicon extraction. However, the datasets are imbalanced in different languages. For some languages, the high-quality corpora are scarce or hard to obtain, which limits the research progress. To solve the above problems, we explore a mutual-reinforcement label propagation framework. First, for each language, a label propagation algorithm is applied to a word relation graph, and then a bilingual dictionary is used as a bridge to transfer information between two languages. A key advantage of this model is its ability to make two languages learn from each other and boost each other. The experimental results show that the proposed approach outperforms baseline significantly.

Cross-Language Opinion Lexicon Extraction Using Mutual-Reinforcement Label Propagation

PubMed Central

Lin, Zheng; Tan, Songbo; Liu, Yue; Cheng, Xueqi; Xu, Xueke

2013-01-01

There is a growing interest in automatically building opinion lexicon from sources such as product reviews. Most of these methods depend on abundant external resources such as WordNet, which limits the applicability of these methods. Unsupervised or semi-supervised learning provides an optional solution to multilingual opinion lexicon extraction. However, the datasets are imbalanced in different languages. For some languages, the high-quality corpora are scarce or hard to obtain, which limits the research progress. To solve the above problems, we explore a mutual-reinforcement label propagation framework. First, for each language, a label propagation algorithm is applied to a word relation graph, and then a bilingual dictionary is used as a bridge to transfer information between two languages. A key advantage of this model is its ability to make two languages learn from each other and boost each other. The experimental results show that the proposed approach outperforms baseline significantly. PMID:24260190

Program Helps Simulate Neural Networks

NASA Technical Reports Server (NTRS)

Villarreal, James; Mcintire, Gary

1993-01-01

Neural Network Environment on Transputer System (NNETS) computer program provides users high degree of flexibility in creating and manipulating wide variety of neural-network topologies at processing speeds not found in conventional computing environments. Supports back-propagation and back-propagation-related algorithms. Back-propagation algorithm used is implementation of Rumelhart's generalized delta rule. NNETS developed on INMOS Transputer(R). Predefines back-propagation network, Jordan network, and reinforcement network to assist users in learning and defining own networks. Also enables users to configure other neural-network paradigms from NNETS basic architecture. Small portion of software written in OCCAM(R) language.

Towards a genetics-based adaptive agent to support flight testing

NASA Astrophysics Data System (ADS)

Cribbs, Henry Brown, III

Although the benefits of aircraft simulation have been known since the late 1960s, simulation almost always entails interaction with a human test pilot. This "pilot-in-the-loop" simulation process provides useful evaluative information to the aircraft designer and provides a training tool to the pilot. Emulation of a pilot during the early phases of the aircraft design process might provide designers a useful evaluative tool. Machine learning might emulate a pilot in a simulated aircraft/cockpit setting. Preliminary work in the application of machine learning techniques, such as reinforcement learning, to aircraft maneuvering have shown promise. These studies used simplified interfaces between machine learning agent and the aircraft simulation. The simulations employed low order equivalent system models. High-fidelity aircraft simulations exist, such as the simulations developed by NASA at its Dryden Flight Research Center. To expand the applicational domain of reinforcement learning to aircraft designs, this study presents a series of experiments that examine a reinforcement learning agent in the role of test pilot. The NASA X-31 and F-106 high-fidelity simulations provide realistic aircraft for the agent to maneuver. The approach of the study is to examine an agent possessing a genetic-based, artificial neural network to approximate long-term, expected cost (Bellman value) in a basic maneuvering task. The experiments evaluate different learning methods based on a common feedback function and an identical task. The learning methods evaluated are: Q-learning, Q(lambda)-learning, SARSA learning, and SARSA(lambda) learning. Experimental results indicate that, while prediction error remain quite high, similar, repeatable behaviors occur in both aircraft. Similar behavior exhibits portability of the agent between aircraft with different handling qualities (dynamics). Besides the adaptive behavior aspects of the study, the genetic algorithm used in the agent is shown to play an additive role in the shaping of the artificial neural network to the prediction task.

Efficient collective swimming by harnessing vortices through deep reinforcement learning.

PubMed

Verma, Siddhartha; Novati, Guido; Koumoutsakos, Petros

2018-06-05

Fish in schooling formations navigate complex flow fields replete with mechanical energy in the vortex wakes of their companions. Their schooling behavior has been associated with evolutionary advantages including energy savings, yet the underlying physical mechanisms remain unknown. We show that fish can improve their sustained propulsive efficiency by placing themselves in appropriate locations in the wake of other swimmers and intercepting judiciously their shed vortices. This swimming strategy leads to collective energy savings and is revealed through a combination of high-fidelity flow simulations with a deep reinforcement learning (RL) algorithm. The RL algorithm relies on a policy defined by deep, recurrent neural nets, with long-short-term memory cells, that are essential for capturing the unsteadiness of the two-way interactions between the fish and the vortical flow field. Surprisingly, we find that swimming in-line with a leader is not associated with energetic benefits for the follower. Instead, "smart swimmer(s)" place themselves at off-center positions, with respect to the axis of the leader(s) and deform their body to synchronize with the momentum of the oncoming vortices, thus enhancing their swimming efficiency at no cost to the leader(s). The results confirm that fish may harvest energy deposited in vortices and support the conjecture that swimming in formation is energetically advantageous. Moreover, this study demonstrates that deep RL can produce navigation algorithms for complex unsteady and vortical flow fields, with promising implications for energy savings in autonomous robotic swarms.

Progressive Learning of Topic Modeling Parameters: A Visual Analytics Framework.

PubMed

El-Assady, Mennatallah; Sevastjanova, Rita; Sperrle, Fabian; Keim, Daniel; Collins, Christopher

2018-01-01

Topic modeling algorithms are widely used to analyze the thematic composition of text corpora but remain difficult to interpret and adjust. Addressing these limitations, we present a modular visual analytics framework, tackling the understandability and adaptability of topic models through a user-driven reinforcement learning process which does not require a deep understanding of the underlying topic modeling algorithms. Given a document corpus, our approach initializes two algorithm configurations based on a parameter space analysis that enhances document separability. We abstract the model complexity in an interactive visual workspace for exploring the automatic matching results of two models, investigating topic summaries, analyzing parameter distributions, and reviewing documents. The main contribution of our work is an iterative decision-making technique in which users provide a document-based relevance feedback that allows the framework to converge to a user-endorsed topic distribution. We also report feedback from a two-stage study which shows that our technique results in topic model quality improvements on two independent measures.

Cooperating with machines.

PubMed

Crandall, Jacob W; Oudah, Mayada; Tennom; Ishowo-Oloko, Fatimah; Abdallah, Sherief; Bonnefon, Jean-François; Cebrian, Manuel; Shariff, Azim; Goodrich, Michael A; Rahwan, Iyad

2018-01-16

Since Alan Turing envisioned artificial intelligence, technical progress has often been measured by the ability to defeat humans in zero-sum encounters (e.g., Chess, Poker, or Go). Less attention has been given to scenarios in which human-machine cooperation is beneficial but non-trivial, such as scenarios in which human and machine preferences are neither fully aligned nor fully in conflict. Cooperation does not require sheer computational power, but instead is facilitated by intuition, cultural norms, emotions, signals, and pre-evolved dispositions. Here, we develop an algorithm that combines a state-of-the-art reinforcement-learning algorithm with mechanisms for signaling. We show that this algorithm can cooperate with people and other algorithms at levels that rival human cooperation in a variety of two-player repeated stochastic games. These results indicate that general human-machine cooperation is achievable using a non-trivial, but ultimately simple, set of algorithmic mechanisms.

Application of a model of instrumental conditioning to mobile robot control

NASA Astrophysics Data System (ADS)

Saksida, Lisa M.; Touretzky, D. S.

1997-09-01

Instrumental conditioning is a psychological process whereby an animal learns to associate its actions with their consequences. This type of learning is exploited in animal training techniques such as 'shaping by successive approximations,' which enables trainers to gradually adjust the animal's behavior by giving strategically timed reinforcements. While this is similar in principle to reinforcement learning, the real phenomenon includes many subtle effects not considered in the machine learning literature. In addition, a good deal of domain information is utilized by an animal learning a new task; it does not start from scratch every time it learns a new behavior. For these reasons, it is not surprising that mobile robot learning algorithms have yet to approach the sophistication and robustness of animal learning. A serious attempt to model instrumental learning could prove fruitful for improving machine learning techniques. In the present paper, we develop a computational theory of shaping at a level appropriate for controlling mobile robots. The theory is based on a series of mechanisms for 'behavior editing,' in which pre-existing behaviors, either innate or previously learned, can be dramatically changed in magnitude, shifted in direction, or otherwise manipulated so as to produce new behavioral routines. We have implemented our theory on Amelia, an RWI B21 mobile robot equipped with a gripper and color video camera. We provide results from training Amelia on several tasks, all of which were constructed as variations of one innate behavior, object-pursuit.

Representing Trust in Cognitive Social Simulations

NASA Astrophysics Data System (ADS)

Pollock, Shawn S.; Alt, Jonathan K.; Darken, Christian J.

Trust plays a critical role in communications, strength of relationships, and information processing at the individual and group level. Cognitive social simulations show promise in providing an experimental platform for the examination of social phenomena such as trust formation. This paper describes the initial attempts at representation of trust in a cognitive social simulation using reinforcement learning algorithms centered around a cooperative Public Commodity game within a dynamic social network.

Leader-Follower Output Synchronization of Linear Heterogeneous Systems With Active Leader Using Reinforcement Learning.

PubMed

Yang, Yongliang; Modares, Hamidreza; Wunsch, Donald C; Yin, Yixin

2018-06-01

This paper develops optimal control protocols for the distributed output synchronization problem of leader-follower multiagent systems with an active leader. Agents are assumed to be heterogeneous with different dynamics and dimensions. The desired trajectory is assumed to be preplanned and is generated by the leader. Other follower agents autonomously synchronize to the leader by interacting with each other using a communication network. The leader is assumed to be active in the sense that it has a nonzero control input so that it can act independently and update its control to keep the followers away from possible danger. A distributed observer is first designed to estimate the leader's state and generate the reference signal for each follower. Then, the output synchronization of leader-follower systems with an active leader is formulated as a distributed optimal tracking problem, and inhomogeneous algebraic Riccati equations (AREs) are derived to solve it. The resulting distributed optimal control protocols not only minimize the steady-state error but also optimize the transient response of the agents. An off-policy reinforcement learning algorithm is developed to solve the inhomogeneous AREs online in real time and without requiring any knowledge of the agents' dynamics. Finally, two simulation examples are conducted to illustrate the effectiveness of the proposed algorithm.

TeachEnG: a Teaching Engine for Genomics.

PubMed

Kim, Minji; Kim, Yeonsung; Qian, Lei; Song, Jun S

2017-10-15

Bioinformatics is a rapidly growing field that has emerged from the synergy of computer science, statistics and biology. Given the interdisciplinary nature of bioinformatics, many students from diverse fields struggle with grasping bioinformatic concepts only from classroom lectures. Interactive tools for helping students reinforce their learning would be thus desirable. Here, we present an interactive online educational tool called TeachEnG (acronym for Teaching Engine for Genomics) for reinforcing key concepts in sequence alignment and phylogenetic tree reconstruction. Our instructional games allow students to align sequences by hand, fill out the dynamic programming matrix in the Needleman-Wunsch global sequence alignment algorithm, and reconstruct phylogenetic trees via the maximum parsimony, Unweighted Pair Group Method with Arithmetic mean (UPGMA) and Neighbor-Joining algorithms. With an easily accessible interface and instant visual feedback, TeachEnG will help promote active learning in bioinformatics. TeachEnG is freely available at http://teacheng.illinois.edu. The source code is available from https://github.com/KnowEnG/TeachEnG under the Artistic License 2.0. It is written in JavaScript and compatible with Firefox, Safari, Chrome and Microsoft Edge. songj@illinois.edu. © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com

The Effects of Sensor Performance as Modeled by Signal Detection Theory on the Performance of Reinforcement Learning in a Target Acquisition Task

NASA Astrophysics Data System (ADS)

Quirion, Nate

Unmanned Aerial Systems (UASs) today are fulfilling more roles than ever before. There is a general push to have these systems feature more advanced autonomous capabilities in the near future. To achieve autonomous behavior requires some unique approaches to control and decision making. More advanced versions of these approaches are able to adapt their own behavior and examine their past experiences to increase their future mission performance. To achieve adaptive behavior and decision making capabilities this study used Reinforcement Learning algorithms. In this research the effects of sensor performance, as modeled through Signal Detection Theory (SDT), on the ability of RL algorithms to accomplish a target localization task are examined. Three levels of sensor sensitivity are simulated and compared to the results of the same system using a perfect sensor. To accomplish the target localization task, a hierarchical architecture used two distinct agents. A simulated human operator is assumed to be a perfect decision maker, and is used in the system feedback. An evaluation of the system is performed using multiple metrics, including episodic reward curves and the time taken to locate all targets. Statistical analyses are employed to detect significant differences in the comparison of steady-state behavior of different systems.

Optimal Reward Functions in Distributed Reinforcement Learning

NASA Technical Reports Server (NTRS)

Wolpert, David H.; Tumer, Kagan

2000-01-01

We consider the design of multi-agent systems so as to optimize an overall world utility function when (1) those systems lack centralized communication and control, and (2) each agents runs a distinct Reinforcement Learning (RL) algorithm. A crucial issue in such design problems is to initialize/update each agent's private utility function, so as to induce best possible world utility. Traditional 'team game' solutions to this problem sidestep this issue and simply assign to each agent the world utility as its private utility function. In previous work we used the 'Collective Intelligence' framework to derive a better choice of private utility functions, one that results in world utility performance up to orders of magnitude superior to that ensuing from use of the team game utility. In this paper we extend these results. We derive the general class of private utility functions that both are easy for the individual agents to learn and that, if learned well, result in high world utility. We demonstrate experimentally that using these new utility functions can result in significantly improved performance over that of our previously proposed utility, over and above that previous utility's superiority to the conventional team game utility.

Neural correlates of forward planning in a spatial decision task in humans

PubMed Central

Simon, Dylan Alexander; Daw, Nathaniel D.

2011-01-01

Although reinforcement learning (RL) theories have been influential in characterizing the brain’s mechanisms for reward-guided choice, the predominant temporal difference (TD) algorithm cannot explain many flexible or goal-directed actions that have been demonstrated behaviorally. We investigate such actions by contrasting an RL algorithm that is model-based, in that it relies on learning a map or model of the task and planning within it, to traditional model-free TD learning. To distinguish these approaches in humans, we used fMRI in a continuous spatial navigation task, in which frequent changes to the layout of the maze forced subjects continually to relearn their favored routes, thereby exposing the RL mechanisms employed. We sought evidence for the neural substrates of such mechanisms by comparing choice behavior and BOLD signals to decision variables extracted from simulations of either algorithm. Both choices and value-related BOLD signals in striatum, though most often associated with TD learning, were better explained by the model-based theory. Further, predecessor quantities for the model-based value computation were correlated with BOLD signals in the medial temporal lobe and frontal cortex. These results point to a significant extension of both the computational and anatomical substrates for RL in the brain. PMID:21471389

Classification of amyotrophic lateral sclerosis disease based on convolutional neural network and reinforcement sample learning algorithm.

PubMed

Sengur, Abdulkadir; Akbulut, Yaman; Guo, Yanhui; Bajaj, Varun

2017-12-01

Electromyogram (EMG) signals contain useful information of the neuromuscular diseases like amyotrophic lateral sclerosis (ALS). ALS is a well-known brain disease, which can progressively degenerate the motor neurons. In this paper, we propose a deep learning based method for efficient classification of ALS and normal EMG signals. Spectrogram, continuous wavelet transform (CWT), and smoothed pseudo Wigner-Ville distribution (SPWVD) have been employed for time-frequency (T-F) representation of EMG signals. A convolutional neural network is employed to classify these features. In it, Two convolution layers, two pooling layer, a fully connected layer and a lost function layer is considered in CNN architecture. The CNN architecture is trained with the reinforcement sample learning strategy. The efficiency of the proposed implementation is tested on publicly available EMG dataset. The dataset contains 89 ALS and 133 normal EMG signals with 24 kHz sampling frequency. Experimental results show 96.80% accuracy. The obtained results are also compared with other methods, which show the superiority of the proposed method.

Intelligent control and cooperation for mobile robots

NASA Astrophysics Data System (ADS)

Stingu, Petru Emanuel

The topic discussed in this work addresses the current research being conducted at the Automation & Robotics Research Institute in the areas of UAV quadrotor control and heterogenous multi-vehicle cooperation. Autonomy can be successfully achieved by a robot under the following conditions: the robot has to be able to acquire knowledge about the environment and itself, and it also has to be able to reason under uncertainty. The control system must react quickly to immediate challenges, but also has to slowly adapt and improve based on accumulated knowledge. The major contribution of this work is the transfer of the ADP algorithms from the purely theoretical environment to the complex real-world robotic platforms that work in real-time and in uncontrolled environments. Many solutions are adopted from those present in nature because they have been proven to be close to optimal in very different settings. For the control of a single platform, reinforcement learning algorithms are used to design suboptimal controllers for a class of complex systems that can be conceptually split in local loops with simpler dynamics and relatively weak coupling to the rest of the system. Optimality is enforced by having a global critic but the curse of dimensionality is avoided by using local actors and intelligent pre-processing of the information used for learning the optimal controllers. The system model is used for constructing the structure of the control system, but on top of that the adaptive neural networks that form the actors use the knowledge acquired during normal operation to get closer to optimal control. In real-world experiments, efficient learning is a strong requirement for success. This is accomplished by using an approximation of the system model to focus the learning for equivalent configurations of the state space. Due to the availability of only local data for training, neural networks with local activation functions are implemented. For the control of a formation of robots subjected to dynamic communication constraints, game theory is used in addition to reinforcement learning. The nodes maintain an extra set of state variables about all the other nodes that they can communicate to. The more important are trust and predictability. They are a way to incorporate knowledge acquired in the past into the control decisions taken by each node. The trust variable provides a simple mechanism for the implementation of reinforcement learning. For robot formations, potential field based control algorithms are used to generate the control commands. The formation structure changes due to the environment and due to the decisions of the nodes. It is a problem of building a graph and coalitions by having distributed decisions but still reaching an optimal behavior globally.

A Reinforcement Sensor Embedded Vertical Handoff Controller for Vehicular Heterogeneous Wireless Networks

PubMed Central

Li, Limin; Xu, Yubin; Soong, Boon-Hee; Ma, Lin

2013-01-01

Vehicular communication platforms that provide real-time access to wireless networks have drawn more and more attention in recent years. IEEE 802.11p is the main radio access technology that supports communication for high mobility terminals, however, due to its limited coverage, IEEE 802.11p is usually deployed by coupling with cellular networks to achieve seamless mobility. In a heterogeneous cellular/802.11p network, vehicular communication is characterized by its short time span in association with a wireless local area network (WLAN). Moreover, for the media access control (MAC) scheme used for WLAN, the network throughput dramatically decreases with increasing user quantity. In response to these compelling problems, we propose a reinforcement sensor (RFS) embedded vertical handoff control strategy to support mobility management. The RFS has online learning capability and can provide optimal handoff decisions in an adaptive fashion without prior knowledge. The algorithm integrates considerations including vehicular mobility, traffic load, handoff latency, and network status. Simulation results verify that the proposed algorithm can adaptively adjust the handoff strategy, allowing users to stay connected to the best network. Furthermore, the algorithm can ensure that RSUs are adequate, thereby guaranteeing a high quality user experience. PMID:24193101

Machine learning algorithms for the creation of clinical healthcare enterprise systems

NASA Astrophysics Data System (ADS)

Mandal, Indrajit

2017-10-01

Clinical recommender systems are increasingly becoming popular for improving modern healthcare systems. Enterprise systems are persuasively used for creating effective nurse care plans to provide nurse training, clinical recommendations and clinical quality control. A novel design of a reliable clinical recommender system based on multiple classifier system (MCS) is implemented. A hybrid machine learning (ML) ensemble based on random subspace method and random forest is presented. The performance accuracy and robustness of proposed enterprise architecture are quantitatively estimated to be above 99% and 97%, respectively (above 95% confidence interval). The study then extends to experimental analysis of the clinical recommender system with respect to the noisy data environment. The ranking of items in nurse care plan is demonstrated using machine learning algorithms (MLAs) to overcome the drawback of the traditional association rule method. The promising experimental results are compared against the sate-of-the-art approaches to highlight the advancement in recommendation technology. The proposed recommender system is experimentally validated using five benchmark clinical data to reinforce the research findings.

Stochastic abstract policies: generalizing knowledge to improve reinforcement learning.

PubMed

Koga, Marcelo L; Freire, Valdinei; Costa, Anna H R

2015-01-01

Reinforcement learning (RL) enables an agent to learn behavior by acquiring experience through trial-and-error interactions with a dynamic environment. However, knowledge is usually built from scratch and learning to behave may take a long time. Here, we improve the learning performance by leveraging prior knowledge; that is, the learner shows proper behavior from the beginning of a target task, using the knowledge from a set of known, previously solved, source tasks. In this paper, we argue that building stochastic abstract policies that generalize over past experiences is an effective way to provide such improvement and this generalization outperforms the current practice of using a library of policies. We achieve that contributing with a new algorithm, AbsProb-PI-multiple and a framework for transferring knowledge represented as a stochastic abstract policy in new RL tasks. Stochastic abstract policies offer an effective way to encode knowledge because the abstraction they provide not only generalizes solutions but also facilitates extracting the similarities among tasks. We perform experiments in a robotic navigation environment and analyze the agent's behavior throughout the learning process and also assess the transfer ratio for different amounts of source tasks. We compare our method with the transfer of a library of policies, and experiments show that the use of a generalized policy produces better results by more effectively guiding the agent when learning a target task.

Deep Reinforcement Learning of Cell Movement in the Early Stage of C. elegans Embryogenesis.

PubMed

Wang, Zi; Wang, Dali; Li, Chengcheng; Xu, Yichi; Li, Husheng; Bao, Zhirong

2018-04-25

Cell movement in the early phase of C. elegans development is regulated by a highly complex process in which a set of rules and connections are formulated at distinct scales. Previous efforts have demonstrated that agent-based, multi-scale modeling systems can integrate physical and biological rules and provide new avenues to study developmental systems. However, the application of these systems to model cell movement is still challenging and requires a comprehensive understanding of regulatory networks at the right scales. Recent developments in deep learning and reinforcement learning provide an unprecedented opportunity to explore cell movement using 3D time-lapse microscopy images. We present a deep reinforcement learning approach within an agent-based modeling system to characterize cell movement in the embryonic development of C. elegans. Our modeling system captures the complexity of cell movement patterns in the embryo and overcomes the local optimization problem encountered by traditional rule-based, agent-based modeling that uses greedy algorithms. We tested our model with two real developmental processes: the anterior movement of the Cpaaa cell via intercalation and the rearrangement of the superficial left-right asymmetry. In the first case, the model results suggested that Cpaaa's intercalation is an active directional cell movement caused by the continuous effects from a longer distance (farther than the length of two adjacent cells), as opposed to a passive movement caused by neighbor cell movements. In the second case, a leader-follower mechanism well explained the collective cell movement pattern in the asymmetry rearrangement. These results showed that our approach to introduce deep reinforcement learning into agent-based modeling can test regulatory mechanisms by exploring cell migration paths in a reverse engineering perspective. This model opens new doors to explore the large datasets generated by live imaging. Source code is available at https://github.com/zwang84/drl4cellmovement. dwang7@utk.edu, baoz@mskcc.org. Supplementary data are available at Bioinformatics online.

Brain-Machine Interface control of a robot arm using actor-critic rainforcement learning.

PubMed

Pohlmeyer, Eric A; Mahmoudi, Babak; Geng, Shijia; Prins, Noeline; Sanchez, Justin C

2012-01-01

Here we demonstrate how a marmoset monkey can use a reinforcement learning (RL) Brain-Machine Interface (BMI) to effectively control the movements of a robot arm for a reaching task. In this work, an actor-critic RL algorithm used neural ensemble activity in the monkey's motor cortext to control the robot movements during a two-target decision task. This novel approach to decoding offers unique advantages for BMI control applications. Compared to supervised learning decoding methods, the actor-critic RL algorithm does not require an explicit set of training data to create a static control model, but rather it incrementally adapts the model parameters according to its current performance, in this case requiring only a very basic feedback signal. We show how this algorithm achieved high performance when mapping the monkey's neural states (94%) to robot actions, and only needed to experience a few trials before obtaining accurate real-time control of the robot arm. Since RL methods responsively adapt and adjust their parameters, they can provide a method to create BMIs that are robust against perturbations caused by changes in either the neural input space or the output actions they generate under different task requirements or goals.

An approximate dynamic programming approach to resource management in multi-cloud scenarios

NASA Astrophysics Data System (ADS)

Pietrabissa, Antonio; Priscoli, Francesco Delli; Di Giorgio, Alessandro; Giuseppi, Alessandro; Panfili, Martina; Suraci, Vincenzo

2017-03-01

The programmability and the virtualisation of network resources are crucial to deploy scalable Information and Communications Technology (ICT) services. The increasing demand of cloud services, mainly devoted to the storage and computing, requires a new functional element, the Cloud Management Broker (CMB), aimed at managing multiple cloud resources to meet the customers' requirements and, simultaneously, to optimise their usage. This paper proposes a multi-cloud resource allocation algorithm that manages the resource requests with the aim of maximising the CMB revenue over time. The algorithm is based on Markov decision process modelling and relies on reinforcement learning techniques to find online an approximate solution.

Rational and Mechanistic Perspectives on Reinforcement Learning

ERIC Educational Resources Information Center

Chater, Nick

2009-01-01

This special issue describes important recent developments in applying reinforcement learning models to capture neural and cognitive function. But reinforcement learning, as a theoretical framework, can apply at two very different levels of description: "mechanistic" and "rational." Reinforcement learning is often viewed in mechanistic terms--as…

A simple computational algorithm of model-based choice preference.

PubMed

Toyama, Asako; Katahira, Kentaro; Ohira, Hideki

2017-08-01

A broadly used computational framework posits that two learning systems operate in parallel during the learning of choice preferences-namely, the model-free and model-based reinforcement-learning systems. In this study, we examined another possibility, through which model-free learning is the basic system and model-based information is its modulator. Accordingly, we proposed several modified versions of a temporal-difference learning model to explain the choice-learning process. Using the two-stage decision task developed by Daw, Gershman, Seymour, Dayan, and Dolan (2011), we compared their original computational model, which assumes a parallel learning process, and our proposed models, which assume a sequential learning process. Choice data from 23 participants showed a better fit with the proposed models. More specifically, the proposed eligibility adjustment model, which assumes that the environmental model can weight the degree of the eligibility trace, can explain choices better under both model-free and model-based controls and has a simpler computational algorithm than the original model. In addition, the forgetting learning model and its variation, which assume changes in the values of unchosen actions, substantially improved the fits to the data. Overall, we show that a hybrid computational model best fits the data. The parameters used in this model succeed in capturing individual tendencies with respect to both model use in learning and exploration behavior. This computational model provides novel insights into learning with interacting model-free and model-based components.

Acquisition of Robotic Giant-swing Motion Using Reinforcement Learning and Its Consideration of Motion Forms

NASA Astrophysics Data System (ADS)

Sakai, Naoki; Kawabe, Naoto; Hara, Masayuki; Toyoda, Nozomi; Yabuta, Tetsuro

This paper argues how a compact humanoid robot can acquire a giant-swing motion without any robotic models by using Q-Learning method. Generally, it is widely said that Q-Learning is not appropriated for learning dynamic motions because Markov property is not necessarily guaranteed during the dynamic task. However, we tried to solve this problem by embedding the angular velocity state into state definition and averaging Q-Learning method to reduce dynamic effects, although there remain non-Markov effects in the learning results. The result shows how the robot can acquire a giant-swing motion by using Q-Learning algorithm. The successful acquired motions are analyzed in the view point of dynamics in order to realize a functionally giant-swing motion. Finally, the result shows how this method can avoid the stagnant action loop at around the bottom of the horizontal bar during the early stage of giant-swing motion.

Learning relative values in the striatum induces violations of normative decision making

PubMed Central

Klein, Tilmann A.; Ullsperger, Markus; Jocham, Gerhard

2017-01-01

To decide optimally between available options, organisms need to learn the values associated with these options. Reinforcement learning models offer a powerful explanation of how these values are learnt from experience. However, human choices often violate normative principles. We suggest that seemingly counterintuitive decisions may arise as a natural consequence of the learning mechanisms deployed by humans. Here, using fMRI and a novel behavioural task, we show that, when suddenly switched to novel choice contexts, participants’ choices are incongruent with values learnt by standard learning algorithms. Instead, behaviour is compatible with the decisions of an agent learning how good an option is relative to an option with which it had previously been paired. Striatal activity exhibits the characteristics of a prediction error used to update such relative option values. Our data suggest that choices can be biased by a tendency to learn option values with reference to the available alternatives. PMID:28631734

A self-taught artificial agent for multi-physics computational model personalization.

PubMed

Neumann, Dominik; Mansi, Tommaso; Itu, Lucian; Georgescu, Bogdan; Kayvanpour, Elham; Sedaghat-Hamedani, Farbod; Amr, Ali; Haas, Jan; Katus, Hugo; Meder, Benjamin; Steidl, Stefan; Hornegger, Joachim; Comaniciu, Dorin

2016-12-01

Personalization is the process of fitting a model to patient data, a critical step towards application of multi-physics computational models in clinical practice. Designing robust personalization algorithms is often a tedious, time-consuming, model- and data-specific process. We propose to use artificial intelligence concepts to learn this task, inspired by how human experts manually perform it. The problem is reformulated in terms of reinforcement learning. In an off-line phase, Vito, our self-taught artificial agent, learns a representative decision process model through exploration of the computational model: it learns how the model behaves under change of parameters. The agent then automatically learns an optimal strategy for on-line personalization. The algorithm is model-independent; applying it to a new model requires only adjusting few hyper-parameters of the agent and defining the observations to match. The full knowledge of the model itself is not required. Vito was tested in a synthetic scenario, showing that it could learn how to optimize cost functions generically. Then Vito was applied to the inverse problem of cardiac electrophysiology and the personalization of a whole-body circulation model. The obtained results suggested that Vito could achieve equivalent, if not better goodness of fit than standard methods, while being more robust (up to 11% higher success rates) and with faster (up to seven times) convergence rate. Our artificial intelligence approach could thus make personalization algorithms generalizable and self-adaptable to any patient and any model. Copyright © 2016. Published by Elsevier B.V.

A game theoretic framework for incentive-based models of intrinsic motivation in artificial systems

PubMed Central

Merrick, Kathryn E.; Shafi, Kamran

2013-01-01

An emerging body of research is focusing on understanding and building artificial systems that can achieve open-ended development influenced by intrinsic motivations. In particular, research in robotics and machine learning is yielding systems and algorithms with increasing capacity for self-directed learning and autonomy. Traditional software architectures and algorithms are being augmented with intrinsic motivations to drive cumulative acquisition of knowledge and skills. Intrinsic motivations have recently been considered in reinforcement learning, active learning and supervised learning settings among others. This paper considers game theory as a novel setting for intrinsic motivation. A game theoretic framework for intrinsic motivation is formulated by introducing the concept of optimally motivating incentive as a lens through which players perceive a game. Transformations of four well-known mixed-motive games are presented to demonstrate the perceived games when players' optimally motivating incentive falls in three cases corresponding to strong power, affiliation and achievement motivation. We use agent-based simulations to demonstrate that players with different optimally motivating incentive act differently as a result of their altered perception of the game. We discuss the implications of these results both for modeling human behavior and for designing artificial agents or robots. PMID:24198797

A game theoretic framework for incentive-based models of intrinsic motivation in artificial systems.

PubMed

Merrick, Kathryn E; Shafi, Kamran

2013-01-01

An emerging body of research is focusing on understanding and building artificial systems that can achieve open-ended development influenced by intrinsic motivations. In particular, research in robotics and machine learning is yielding systems and algorithms with increasing capacity for self-directed learning and autonomy. Traditional software architectures and algorithms are being augmented with intrinsic motivations to drive cumulative acquisition of knowledge and skills. Intrinsic motivations have recently been considered in reinforcement learning, active learning and supervised learning settings among others. This paper considers game theory as a novel setting for intrinsic motivation. A game theoretic framework for intrinsic motivation is formulated by introducing the concept of optimally motivating incentive as a lens through which players perceive a game. Transformations of four well-known mixed-motive games are presented to demonstrate the perceived games when players' optimally motivating incentive falls in three cases corresponding to strong power, affiliation and achievement motivation. We use agent-based simulations to demonstrate that players with different optimally motivating incentive act differently as a result of their altered perception of the game. We discuss the implications of these results both for modeling human behavior and for designing artificial agents or robots.

Interactive machine learning for health informatics: when do we need the human-in-the-loop?

PubMed

Holzinger, Andreas

2016-06-01

Machine learning (ML) is the fastest growing field in computer science, and health informatics is among the greatest challenges. The goal of ML is to develop algorithms which can learn and improve over time and can be used for predictions. Most ML researchers concentrate on automatic machine learning (aML), where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from big data with many training sets. However, in the health domain, sometimes we are confronted with a small number of data sets or rare events, where aML-approaches suffer of insufficient training samples. Here interactive machine learning (iML) may be of help, having its roots in reinforcement learning, preference learning, and active learning. The term iML is not yet well used, so we define it as "algorithms that can interact with agents and can optimize their learning behavior through these interactions, where the agents can also be human." This "human-in-the-loop" can be beneficial in solving computationally hard problems, e.g., subspace clustering, protein folding, or k-anonymization of health data, where human expertise can help to reduce an exponential search space through heuristic selection of samples. Therefore, what would otherwise be an NP-hard problem, reduces greatly in complexity through the input and the assistance of a human agent involved in the learning phase.

Negative reinforcement learning is affected in substance dependence.

PubMed

Thompson, Laetitia L; Claus, Eric D; Mikulich-Gilbertson, Susan K; Banich, Marie T; Crowley, Thomas; Krmpotich, Theodore; Miller, David; Tanabe, Jody

2012-06-01

Negative reinforcement results in behavior to escape or avoid an aversive outcome. Withdrawal symptoms are purported to be negative reinforcers in perpetuating substance dependence, but little is known about negative reinforcement learning in this population. The purpose of this study was to examine reinforcement learning in substance dependent individuals (SDI), with an emphasis on assessing negative reinforcement learning. We modified the Iowa Gambling Task to separately assess positive and negative reinforcement. We hypothesized that SDI would show differences in negative reinforcement learning compared to controls and we investigated whether learning differed as a function of the relative magnitude or frequency of the reinforcer. Thirty subjects dependent on psychostimulants were compared with 28 community controls on a decision making task that manipulated outcome frequencies and magnitudes and required an action to avoid a negative outcome. SDI did not learn to avoid negative outcomes to the same degree as controls. This difference was driven by the magnitude, not the frequency, of negative feedback. In contrast, approach behaviors in response to positive reinforcement were similar in both groups. Our findings are consistent with a specific deficit in negative reinforcement learning in SDI. SDI were relatively insensitive to the magnitude, not frequency, of loss. If this generalizes to drug-related stimuli, it suggests that repeated episodes of withdrawal may drive relapse more than the severity of a single episode. Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.

An Upside to Reward Sensitivity: The Hippocampus Supports Enhanced Reinforcement Learning in Adolescence.

PubMed

Davidow, Juliet Y; Foerde, Karin; Galván, Adriana; Shohamy, Daphna

2016-10-05

Adolescents are notorious for engaging in reward-seeking behaviors, a tendency attributed to heightened activity in the brain's reward systems during adolescence. It has been suggested that reward sensitivity in adolescence might be adaptive, but evidence of an adaptive role has been scarce. Using a probabilistic reinforcement learning task combined with reinforcement learning models and fMRI, we found that adolescents showed better reinforcement learning and a stronger link between reinforcement learning and episodic memory for rewarding outcomes. This behavioral benefit was related to heightened prediction error-related BOLD activity in the hippocampus and to stronger functional connectivity between the hippocampus and the striatum at the time of reinforcement. These findings reveal an important role for the hippocampus in reinforcement learning in adolescence and suggest that reward sensitivity in adolescence is related to adaptive differences in how adolescents learn from experience. Copyright © 2016 Elsevier Inc. All rights reserved.

Statistical Mechanics of Node-perturbation Learning with Noisy Baseline

NASA Astrophysics Data System (ADS)

Hara, Kazuyuki; Katahira, Kentaro; Okada, Masato

2017-02-01

Node-perturbation learning is a type of statistical gradient descent algorithm that can be applied to problems where the objective function is not explicitly formulated, including reinforcement learning. It estimates the gradient of an objective function by using the change in the object function in response to the perturbation. The value of the objective function for an unperturbed output is called a baseline. Cho et al. proposed node-perturbation learning with a noisy baseline. In this paper, we report on building the statistical mechanics of Cho's model and on deriving coupled differential equations of order parameters that depict learning dynamics. We also show how to derive the generalization error by solving the differential equations of order parameters. On the basis of the results, we show that Cho's results are also apply in general cases and show some general performances of Cho's model.

Model-Free Machine Learning in Biomedicine: Feasibility Study in Type 1 Diabetes

PubMed Central

Daskalaki, Elena; Diem, Peter; Mougiakakou, Stavroula G.

2016-01-01

Although reinforcement learning (RL) is suitable for highly uncertain systems, the applicability of this class of algorithms to medical treatment may be limited by the patient variability which dictates individualised tuning for their usually multiple algorithmic parameters. This study explores the feasibility of RL in the framework of artificial pancreas development for type 1 diabetes (T1D). In this approach, an Actor-Critic (AC) learning algorithm is designed and developed for the optimisation of insulin infusion for personalised glucose regulation. AC optimises the daily basal insulin rate and insulin:carbohydrate ratio for each patient, on the basis of his/her measured glucose profile. Automatic, personalised tuning of AC is based on the estimation of information transfer (IT) from insulin to glucose signals. Insulin-to-glucose IT is linked to patient-specific characteristics related to total daily insulin needs and insulin sensitivity (SI). The AC algorithm is evaluated using an FDA-accepted T1D simulator on a large patient database under a complex meal protocol, meal uncertainty and diurnal SI variation. The results showed that 95.66% of time was spent in normoglycaemia in the presence of meal uncertainty and 93.02% when meal uncertainty and SI variation were simultaneously considered. The time spent in hypoglycaemia was 0.27% in both cases. The novel tuning method reduced the risk of severe hypoglycaemia, especially in patients with low SI. PMID:27441367

Bayesian Cue Integration as a Developmental Outcome of Reward Mediated Learning

PubMed Central

Weisswange, Thomas H.; Rothkopf, Constantin A.; Rodemann, Tobias; Triesch, Jochen

2011-01-01

Average human behavior in cue combination tasks is well predicted by Bayesian inference models. As this capability is acquired over developmental timescales, the question arises, how it is learned. Here we investigated whether reward dependent learning, that is well established at the computational, behavioral, and neuronal levels, could contribute to this development. It is shown that a model free reinforcement learning algorithm can indeed learn to do cue integration, i.e. weight uncertain cues according to their respective reliabilities and even do so if reliabilities are changing. We also consider the case of causal inference where multimodal signals can originate from one or multiple separate objects and should not always be integrated. In this case, the learner is shown to develop a behavior that is closest to Bayesian model averaging. We conclude that reward mediated learning could be a driving force for the development of cue integration and causal inference. PMID:21750717

Variability in Dopamine Genes Dissociates Model-Based and Model-Free Reinforcement Learning

PubMed Central

Bath, Kevin G.; Daw, Nathaniel D.; Frank, Michael J.

2016-01-01

Considerable evidence suggests that multiple learning systems can drive behavior. Choice can proceed reflexively from previous actions and their associated outcomes, as captured by “model-free” learning algorithms, or flexibly from prospective consideration of outcomes that might occur, as captured by “model-based” learning algorithms. However, differential contributions of dopamine to these systems are poorly understood. Dopamine is widely thought to support model-free learning by modulating plasticity in striatum. Model-based learning may also be affected by these striatal effects, or by other dopaminergic effects elsewhere, notably on prefrontal working memory function. Indeed, prominent demonstrations linking striatal dopamine to putatively model-free learning did not rule out model-based effects, whereas other studies have reported dopaminergic modulation of verifiably model-based learning, but without distinguishing a prefrontal versus striatal locus. To clarify the relationships between dopamine, neural systems, and learning strategies, we combine a genetic association approach in humans with two well-studied reinforcement learning tasks: one isolating model-based from model-free behavior and the other sensitive to key aspects of striatal plasticity. Prefrontal function was indexed by a polymorphism in the COMT gene, differences of which reflect dopamine levels in the prefrontal cortex. This polymorphism has been associated with differences in prefrontal activity and working memory. Striatal function was indexed by a gene coding for DARPP-32, which is densely expressed in the striatum where it is necessary for synaptic plasticity. We found evidence for our hypothesis that variations in prefrontal dopamine relate to model-based learning, whereas variations in striatal dopamine function relate to model-free learning. SIGNIFICANCE STATEMENT Decisions can stem reflexively from their previously associated outcomes or flexibly from deliberative consideration of potential choice outcomes. Research implicates a dopamine-dependent striatal learning mechanism in the former type of choice. Although recent work has indicated that dopamine is also involved in flexible, goal-directed decision-making, it remains unclear whether it also contributes via striatum or via the dopamine-dependent working memory function of prefrontal cortex. We examined genetic indices of dopamine function in these regions and their relation to the two choice strategies. We found that striatal dopamine function related most clearly to the reflexive strategy, as previously shown, and that prefrontal dopamine related most clearly to the flexible strategy. These findings suggest that dissociable brain regions support dissociable choice strategies. PMID:26818509

Spatial-Temporal Reasoning Applications of Computational Intelligence in the Game of Go and Computer Networks

DTIC Science & Technology

2012-01-01

dimensionality, Tesauro used a backpropagation- based , three-layer neural network and implemented the outcome from a self-play game as the reinforcement signal...a school of fish, flock of birds, and colony of ants. Our literature review reveals that no one has used PSO to train the neural network ...trained with a variant of PSO called cellular PSO (CPSO). CSRN is a supervised learning neural network (SLNN). The proposed algorithm for the

Roles of OA1 octopamine receptor and Dop1 dopamine receptor in mediating appetitive and aversive reinforcement revealed by RNAi studies

PubMed Central

Awata, Hiroko; Wakuda, Ryo; Ishimaru, Yoshiyasu; Matsuoka, Yuji; Terao, Kanta; Katata, Satomi; Matsumoto, Yukihisa; Hamanaka, Yoshitaka; Noji, Sumihare; Mito, Taro; Mizunami, Makoto

2016-01-01

Revealing reinforcing mechanisms in associative learning is important for elucidation of brain mechanisms of behavior. In mammals, dopamine neurons are thought to mediate both appetitive and aversive reinforcement signals. Studies using transgenic fruit-flies suggested that dopamine neurons mediate both appetitive and aversive reinforcements, through the Dop1 dopamine receptor, but our studies using octopamine and dopamine receptor antagonists and using Dop1 knockout crickets suggested that octopamine neurons mediate appetitive reinforcement and dopamine neurons mediate aversive reinforcement in associative learning in crickets. To fully resolve this issue, we examined the effects of silencing of expression of genes that code the OA1 octopamine receptor and Dop1 and Dop2 dopamine receptors by RNAi in crickets. OA1-silenced crickets exhibited impairment in appetitive learning with water but not in aversive learning with sodium chloride solution, while Dop1-silenced crickets exhibited impairment in aversive learning but not in appetitive learning. Dop2-silenced crickets showed normal scores in both appetitive learning and aversive learning. The results indicate that octopamine neurons mediate appetitive reinforcement via OA1 and that dopamine neurons mediate aversive reinforcement via Dop1 in crickets, providing decisive evidence that neurotransmitters and receptors that mediate appetitive reinforcement indeed differ among different species of insects. PMID:27412401

Roles of OA1 octopamine receptor and Dop1 dopamine receptor in mediating appetitive and aversive reinforcement revealed by RNAi studies.

PubMed

Awata, Hiroko; Wakuda, Ryo; Ishimaru, Yoshiyasu; Matsuoka, Yuji; Terao, Kanta; Katata, Satomi; Matsumoto, Yukihisa; Hamanaka, Yoshitaka; Noji, Sumihare; Mito, Taro; Mizunami, Makoto

2016-07-14

Revealing reinforcing mechanisms in associative learning is important for elucidation of brain mechanisms of behavior. In mammals, dopamine neurons are thought to mediate both appetitive and aversive reinforcement signals. Studies using transgenic fruit-flies suggested that dopamine neurons mediate both appetitive and aversive reinforcements, through the Dop1 dopamine receptor, but our studies using octopamine and dopamine receptor antagonists and using Dop1 knockout crickets suggested that octopamine neurons mediate appetitive reinforcement and dopamine neurons mediate aversive reinforcement in associative learning in crickets. To fully resolve this issue, we examined the effects of silencing of expression of genes that code the OA1 octopamine receptor and Dop1 and Dop2 dopamine receptors by RNAi in crickets. OA1-silenced crickets exhibited impairment in appetitive learning with water but not in aversive learning with sodium chloride solution, while Dop1-silenced crickets exhibited impairment in aversive learning but not in appetitive learning. Dop2-silenced crickets showed normal scores in both appetitive learning and aversive learning. The results indicate that octopamine neurons mediate appetitive reinforcement via OA1 and that dopamine neurons mediate aversive reinforcement via Dop1 in crickets, providing decisive evidence that neurotransmitters and receptors that mediate appetitive reinforcement indeed differ among different species of insects.

Approximately adaptive neural cooperative control for nonlinear multiagent systems with performance guarantee

NASA Astrophysics Data System (ADS)

Wang, Jing; Yang, Tianyu; Staskevich, Gennady; Abbe, Brian

2017-04-01

This paper studies the cooperative control problem for a class of multiagent dynamical systems with partially unknown nonlinear system dynamics. In particular, the control objective is to solve the state consensus problem for multiagent systems based on the minimisation of certain cost functions for individual agents. Under the assumption that there exist admissible cooperative controls for such class of multiagent systems, the formulated problem is solved through finding the optimal cooperative control using the approximate dynamic programming and reinforcement learning approach. With the aid of neural network parameterisation and online adaptive learning, our method renders a practically implementable approximately adaptive neural cooperative control for multiagent systems. Specifically, based on the Bellman's principle of optimality, the Hamilton-Jacobi-Bellman (HJB) equation for multiagent systems is first derived. We then propose an approximately adaptive policy iteration algorithm for multiagent cooperative control based on neural network approximation of the value functions. The convergence of the proposed algorithm is rigorously proved using the contraction mapping method. The simulation results are included to validate the effectiveness of the proposed algorithm.

Longitudinal investigation on learned helplessness tested under negative and positive reinforcement involving stimulus control.

PubMed

Oliveira, Emileane C; Hunziker, Maria Helena

2014-07-01

In this study, we investigated whether (a) animals demonstrating the learned helplessness effect during an escape contingency also show learning deficits under positive reinforcement contingencies involving stimulus control and (b) the exposure to positive reinforcement contingencies eliminates the learned helplessness effect under an escape contingency. Rats were initially exposed to controllable (C), uncontrollable (U) or no (N) shocks. After 24h, they were exposed to 60 escapable shocks delivered in a shuttlebox. In the following phase, we selected from each group the four subjects that presented the most typical group pattern: no escape learning (learned helplessness effect) in Group U and escape learning in Groups C and N. All subjects were then exposed to two phases, the (1) positive reinforcement for lever pressing under a multiple FR/Extinction schedule and (2) a re-test under negative reinforcement (escape). A fourth group (n=4) was exposed only to the positive reinforcement sessions. All subjects showed discrimination learning under multiple schedule. In the escape re-test, the learned helplessness effect was maintained for three of the animals in Group U. These results suggest that the learned helplessness effect did not extend to discriminative behavior that is positively reinforced and that the learned helplessness effect did not revert for most subjects after exposure to positive reinforcement. We discuss some theoretical implications as related to learned helplessness as an effect restricted to aversive contingencies and to the absence of reversion after positive reinforcement. This article is part of a Special Issue entitled: insert SI title. Copyright © 2014. Published by Elsevier B.V.

Rational and mechanistic perspectives on reinforcement learning.

PubMed

Chater, Nick

2009-12-01

This special issue describes important recent developments in applying reinforcement learning models to capture neural and cognitive function. But reinforcement learning, as a theoretical framework, can apply at two very different levels of description: mechanistic and rational. Reinforcement learning is often viewed in mechanistic terms--as describing the operation of aspects of an agent's cognitive and neural machinery. Yet it can also be viewed as a rational level of description, specifically, as describing a class of methods for learning from experience, using minimal background knowledge. This paper considers how rational and mechanistic perspectives differ, and what types of evidence distinguish between them. Reinforcement learning research in the cognitive and brain sciences is often implicitly committed to the mechanistic interpretation. Here the opposite view is put forward: that accounts of reinforcement learning should apply at the rational level, unless there is strong evidence for a mechanistic interpretation. Implications of this viewpoint for reinforcement-based theories in the cognitive and brain sciences are discussed.

Stress affects instrumental learning based on positive or negative reinforcement in interaction with personality in domestic horses

PubMed Central

Valenchon, Mathilde; Lévy, Frédéric; Moussu, Chantal; Lansade, Léa

2017-01-01

The present study investigated how stress affects instrumental learning performance in horses (Equus caballus) depending on the type of reinforcement. Horses were assigned to four groups (N = 15 per group); each group received training with negative or positive reinforcement in the presence or absence of stressors unrelated to the learning task. The instrumental learning task consisted of the horse entering one of two compartments at the appearance of a visual signal given by the experimenter. In the absence of stressors unrelated to the task, learning performance did not differ between negative and positive reinforcements. The presence of stressors unrelated to the task (exposure to novel and sudden stimuli) impaired learning performance. Interestingly, this learning deficit was smaller when the negative reinforcement was used. The negative reinforcement, considered as a stressor related to the task, could have counterbalanced the impact of the extrinsic stressor by focusing attention toward the learning task. In addition, learning performance appears to differ between certain dimensions of personality depending on the presence of stressors and the type of reinforcement. These results suggest that when negative reinforcement is used (i.e. stressor related to the task), the most fearful horses may be the best performers in the absence of stressors but the worst performers when stressors are present. On the contrary, when positive reinforcement is used, the most fearful horses appear to be consistently the worst performers, with and without exposure to stressors unrelated to the learning task. This study is the first to demonstrate in ungulates that stress affects learning performance differentially according to the type of reinforcement and in interaction with personality. It provides fundamental and applied perspectives in the understanding of the relationships between personality and training abilities. PMID:28475581

Automatic spin-chain learning to explore the quantum speed limit

NASA Astrophysics Data System (ADS)

Zhang, Xiao-Ming; Cui, Zi-Wei; Wang, Xin; Yung, Man-Hong

2018-05-01

One of the ambitious goals of artificial intelligence is to build a machine that outperforms human intelligence, even if limited knowledge and data are provided. Reinforcement learning (RL) provides one such possibility to reach this goal. In this work, we consider a specific task from quantum physics, i.e., quantum state transfer in a one-dimensional spin chain. The mission for the machine is to find transfer schemes with the fastest speeds while maintaining high transfer fidelities. The first scenario we consider is when the Hamiltonian is time independent. We update the coupling strength by minimizing a loss function dependent on both the fidelity and the speed. Compared with a scheme proven to be at the quantum speed limit for the perfect state transfer, the scheme provided by RL is faster while maintaining the infidelity below 5 ×10-4 . In the second scenario where a time-dependent external field is introduced, we convert the state transfer process into a Markov decision process that can be understood by the machine. We solve it with the deep Q-learning algorithm. After training, the machine successfully finds transfer schemes with high fidelities and speeds, which are faster than previously known ones. These results show that reinforcement learning can be a powerful tool for quantum control problems.

Efficiency of goal-oriented communicating agents in different graph topologies: A study with Internet crawlers

NASA Astrophysics Data System (ADS)

Lőrincz, András; Lázár, Katalin A.; Palotai, Zsolt

2007-05-01

To what extent does the communication make a goal-oriented community efficient in different topologies? In order to gain insight into this problem, we study the influence of learning method as well as that of the topology of the environment on the communication efficiency of crawlers in quest of novel information in different topics on the Internet. Individual crawlers employ selective learning, function approximation-based reinforcement learning (RL), and their combination. Selective learning, in effect, modifies the starting URL lists of the crawlers, whilst RL alters the URL orderings. Real data have been collected from the web and scale-free worlds, scale-free small world (SFSW), and random world environments (RWEs) have been created by link reorganization. In our previous experiments [ Zs. Palotai, Cs. Farkas, A. Lőrincz, Is selection optimal in scale-free small worlds?, ComPlexUs 3 (2006) 158-168], the crawlers searched for novel, genuine documents and direct communication was not possible. Herein, our finding is reproduced: selective learning performs the best and RL the worst in SFSW, whereas the combined, i.e., selective learning coupled with RL is the best-by a slight margin-in scale-free worlds. This effect is demonstrated to be more pronounced when the crawlers search for different topic-specific documents: the relative performance of the combined learning algorithm improves in all worlds, i.e., in SFSW, in SFW, and in RWE. If the tasks are more complex and the work sharing is enforced by the environment then the combined learning algorithm becomes at least equal, even superior to both the selective and the RL algorithms in most cases, irrespective of the efficiency of communication. Furthermore, communication improves the performance by a large margin and adaptive communication is advantageous in the majority of the cases.

Racial bias shapes social reinforcement learning.

PubMed

Lindström, Björn; Selbing, Ida; Molapour, Tanaz; Olsson, Andreas

2014-03-01

Both emotional facial expressions and markers of racial-group belonging are ubiquitous signals in social interaction, but little is known about how these signals together affect future behavior through learning. To address this issue, we investigated how emotional (threatening or friendly) in-group and out-group faces reinforced behavior in a reinforcement-learning task. We asked whether reinforcement learning would be modulated by intergroup attitudes (i.e., racial bias). The results showed that individual differences in racial bias critically modulated reinforcement learning. As predicted, racial bias was associated with more efficiently learned avoidance of threatening out-group individuals. We used computational modeling analysis to quantitatively delimit the underlying processes affected by social reinforcement. These analyses showed that racial bias modulates the rate at which exposure to threatening out-group individuals is transformed into future avoidance behavior. In concert, these results shed new light on the learning processes underlying social interaction with racial-in-group and out-group individuals.

Hierarchically organized behavior and its neural foundations: A reinforcement-learning perspective

PubMed Central

Botvinick, Matthew M.; Niv, Yael; Barto, Andrew C.

2009-01-01

Research on human and animal behavior has long emphasized its hierarchical structure — the divisibility of ongoing behavior into discrete tasks, which are comprised of subtask sequences, which in turn are built of simple actions. The hierarchical structure of behavior has also been of enduring interest within neuroscience, where it has been widely considered to reflect prefrontal cortical functions. In this paper, we reexamine behavioral hierarchy and its neural substrates from the point of view of recent developments in computational reinforcement learning. Specifically, we consider a set of approaches known collectively as hierarchical reinforcement learning, which extend the reinforcement learning paradigm by allowing the learning agent to aggregate actions into reusable subroutines or skills. A close look at the components of hierarchical reinforcement learning suggests how they might map onto neural structures, in particular regions within the dorsolateral and orbital prefrontal cortex. It also suggests specific ways in which hierarchical reinforcement learning might provide a complement to existing psychological models of hierarchically structured behavior. A particularly important question that hierarchical reinforcement learning brings to the fore is that of how learning identifies new action routines that are likely to provide useful building blocks in solving a wide range of future problems. Here and at many other points, hierarchical reinforcement learning offers an appealing framework for investigating the computational and neural underpinnings of hierarchically structured behavior. PMID:18926527

A Policy Representation Using Weighted Multiple Normal Distribution

NASA Astrophysics Data System (ADS)

Kimura, Hajime; Aramaki, Takeshi; Kobayashi, Shigenobu

In this paper, we challenge to solve a reinforcement learning problem for a 5-linked ring robot within a real-time so that the real-robot can stand up to the trial and error. On this robot, incomplete perception problems are caused from noisy sensors and cheap position-control motor systems. This incomplete perception also causes varying optimum actions with the progress of the learning. To cope with this problem, we adopt an actor-critic method, and we propose a new hierarchical policy representation scheme, that consists of discrete action selection on the top level and continuous action selection on the low level of the hierarchy. The proposed hierarchical scheme accelerates learning on continuous action space, and it can pursue the optimum actions varying with the progress of learning on our robotics problem. This paper compares and discusses several learning algorithms through simulations, and demonstrates the proposed method showing application for the real robot.

Design and Control of Large Collections of Learning Agents

NASA Technical Reports Server (NTRS)

Agogino, Adrian

2001-01-01

The intelligent control of multiple autonomous agents is an important yet difficult task. Previous methods used to address this problem have proved to be either too brittle, too hard to use, or not scalable to large systems. The 'Collective Intelligence' project at NASA/Ames provides an elegant, machine-learning approach to address these problems. This approach mathematically defines some essential properties that a reward system should have to promote coordinated behavior among reinforcement learners. This work has focused on creating additional key properties and algorithms within the mathematics of the Collective Intelligence framework. One of the additions will allow agents to learn more quickly, in a more coordinated manner. The other will let agents learn with less knowledge of their environment. These additions will allow the framework to be applied more easily, to a much larger domain of multi-agent problems.

Navigating complex decision spaces: Problems and paradigms in sequential choice

PubMed Central

Walsh, Matthew M.; Anderson, John R.

2015-01-01

To behave adaptively, we must learn from the consequences of our actions. Doing so is difficult when the consequences of an action follow a delay. This introduces the problem of temporal credit assignment. When feedback follows a sequence of decisions, how should the individual assign credit to the intermediate actions that comprise the sequence? Research in reinforcement learning provides two general solutions to this problem: model-free reinforcement learning and model-based reinforcement learning. In this review, we examine connections between stimulus-response and cognitive learning theories, habitual and goal-directed control, and model-free and model-based reinforcement learning. We then consider a range of problems related to temporal credit assignment. These include second-order conditioning and secondary reinforcers, latent learning and detour behavior, partially observable Markov decision processes, actions with distributed outcomes, and hierarchical learning. We ask whether humans and animals, when faced with these problems, behave in a manner consistent with reinforcement learning techniques. Throughout, we seek to identify neural substrates of model-free and model-based reinforcement learning. The former class of techniques is understood in terms of the neurotransmitter dopamine and its effects in the basal ganglia. The latter is understood in terms of a distributed network of regions including the prefrontal cortex, medial temporal lobes cerebellum, and basal ganglia. Not only do reinforcement learning techniques have a natural interpretation in terms of human and animal behavior, but they also provide a useful framework for understanding neural reward valuation and action selection. PMID:23834192

Evolutionary algorithm optimization of biological learning parameters in a biomimetic neuroprosthesis

PubMed Central

Dura-Bernal, S.; Neymotin, S. A.; Kerr, C. C.; Sivagnanam, S.; Majumdar, A.; Francis, J. T.; Lytton, W. W.

2017-01-01

Biomimetic simulation permits neuroscientists to better understand the complex neuronal dynamics of the brain. Embedding a biomimetic simulation in a closed-loop neuroprosthesis, which can read and write signals from the brain, will permit applications for amelioration of motor, psychiatric, and memory-related brain disorders. Biomimetic neuroprostheses require real-time adaptation to changes in the external environment, thus constituting an example of a dynamic data-driven application system. As model fidelity increases, so does the number of parameters and the complexity of finding appropriate parameter configurations. Instead of adapting synaptic weights via machine learning, we employed major biological learning methods: spike-timing dependent plasticity and reinforcement learning. We optimized the learning metaparameters using evolutionary algorithms, which were implemented in parallel and which used an island model approach to obtain sufficient speed. We employed these methods to train a cortical spiking model to utilize macaque brain activity, indicating a selected target, to drive a virtual musculoskeletal arm with realistic anatomical and biomechanical properties to reach to that target. The optimized system was able to reproduce macaque data from a comparable experimental motor task. These techniques can be used to efficiently tune the parameters of multiscale systems, linking realistic neuronal dynamics to behavior, and thus providing a useful tool for neuroscience and neuroprosthetics. PMID:29200477

Distinct prediction errors in mesostriatal circuits of the human brain mediate learning about the values of both states and actions: evidence from high-resolution fMRI.

PubMed

Colas, Jaron T; Pauli, Wolfgang M; Larsen, Tobias; Tyszka, J Michael; O'Doherty, John P

2017-10-01

Prediction-error signals consistent with formal models of "reinforcement learning" (RL) have repeatedly been found within dopaminergic nuclei of the midbrain and dopaminoceptive areas of the striatum. However, the precise form of the RL algorithms implemented in the human brain is not yet well determined. Here, we created a novel paradigm optimized to dissociate the subtypes of reward-prediction errors that function as the key computational signatures of two distinct classes of RL models-namely, "actor/critic" models and action-value-learning models (e.g., the Q-learning model). The state-value-prediction error (SVPE), which is independent of actions, is a hallmark of the actor/critic architecture, whereas the action-value-prediction error (AVPE) is the distinguishing feature of action-value-learning algorithms. To test for the presence of these prediction-error signals in the brain, we scanned human participants with a high-resolution functional magnetic-resonance imaging (fMRI) protocol optimized to enable measurement of neural activity in the dopaminergic midbrain as well as the striatal areas to which it projects. In keeping with the actor/critic model, the SVPE signal was detected in the substantia nigra. The SVPE was also clearly present in both the ventral striatum and the dorsal striatum. However, alongside these purely state-value-based computations we also found evidence for AVPE signals throughout the striatum. These high-resolution fMRI findings suggest that model-free aspects of reward learning in humans can be explained algorithmically with RL in terms of an actor/critic mechanism operating in parallel with a system for more direct action-value learning.

Learning to Control Advanced Life Support Systems

NASA Technical Reports Server (NTRS)

Subramanian, Devika

2004-01-01

Advanced life support systems have many interacting processes and limited resources. Controlling and optimizing advanced life support systems presents unique challenges. In particular, advanced life support systems are nonlinear coupled dynamical systems and it is difficult for humans to take all interactions into account to design an effective control strategy. In this project. we developed several reinforcement learning controllers that actively explore the space of possible control strategies, guided by rewards from a user specified long term objective function. We evaluated these controllers using a discrete event simulation of an advanced life support system. This simulation, called BioSim, designed by Nasa scientists David Kortenkamp and Scott Bell has multiple, interacting life support modules including crew, food production, air revitalization, water recovery, solid waste incineration and power. They are implemented in a consumer/producer relationship in which certain modules produce resources that are consumed by other modules. Stores hold resources between modules. Control of this simulation is via adjusting flows of resources between modules and into/out of stores. We developed adaptive algorithms that control the flow of resources in BioSim. Our learning algorithms discovered several ingenious strategies for maximizing mission length by controlling the air and water recycling systems as well as crop planting schedules. By exploiting non-linearities in the overall system dynamics, the learned controllers easily out- performed controllers written by human experts. In sum, we accomplished three goals. We (1) developed foundations for learning models of coupled dynamical systems by active exploration of the state space, (2) developed and tested algorithms that learn to efficiently control air and water recycling processes as well as crop scheduling in Biosim, and (3) developed an understanding of the role machine learning in designing control systems for advanced life support.

The Role of Multiple Neuromodulators in Reinforcement Learning That Is Based on Competition between Eligibility Traces.

PubMed

Huertas, Marco A; Schwettmann, Sarah E; Shouval, Harel Z

2016-01-01

The ability to maximize reward and avoid punishment is essential for animal survival. Reinforcement learning (RL) refers to the algorithms used by biological or artificial systems to learn how to maximize reward or avoid negative outcomes based on past experiences. While RL is also important in machine learning, the types of mechanistic constraints encountered by biological machinery might be different than those for artificial systems. Two major problems encountered by RL are how to relate a stimulus with a reinforcing signal that is delayed in time (temporal credit assignment), and how to stop learning once the target behaviors are attained (stopping rule). To address the first problem synaptic eligibility traces were introduced, bridging the temporal gap between a stimulus and its reward. Although, these were mere theoretical constructs, recent experiments have provided evidence of their existence. These experiments also reveal that the presence of specific neuromodulators converts the traces into changes in synaptic efficacy. A mechanistic implementation of the stopping rule usually assumes the inhibition of the reward nucleus; however, recent experimental results have shown that learning terminates at the appropriate network state even in setups where the reward nucleus cannot be inhibited. In an effort to describe a learning rule that solves the temporal credit assignment problem and implements a biologically plausible stopping rule, we proposed a model based on two separate synaptic eligibility traces, one for long-term potentiation (LTP) and one for long-term depression (LTD), each obeying different dynamics and having different effective magnitudes. The model has been shown to successfully generate stable learning in recurrent networks. Although, the model assumes the presence of a single neuromodulator, evidence indicates that there are different neuromodulators for expressing the different traces. What could be the role of different neuromodulators for expressing the LTP and LTD traces? Here we expand on our previous model to include several neuromodulators, and illustrate through various examples how different these contribute to learning reward-timing within a wide set of training paradigms and propose further roles that multiple neuromodulators can play in encoding additional information of the rewarding signal.

Effects of dopamine on reinforcement learning and consolidation in Parkinson's disease.

PubMed

Grogan, John P; Tsivos, Demitra; Smith, Laura; Knight, Brogan E; Bogacz, Rafal; Whone, Alan; Coulthard, Elizabeth J

2017-07-10

Emerging evidence suggests that dopamine may modulate learning and memory with important implications for understanding the neurobiology of memory and future therapeutic targeting. An influential hypothesis posits that dopamine biases reinforcement learning. More recent data also suggest an influence during both consolidation and retrieval. Eighteen Parkinson's disease patients learned through feedback ON or OFF medication, with memory tested 24 hr later ON or OFF medication (4 conditions, within-subjects design with matched healthy control group). Patients OFF medication during learning decreased in memory accuracy over the following 24 hr. In contrast to previous studies, however, dopaminergic medication during learning and testing did not affect expression of positive or negative reinforcement. Two further experiments were run without the 24 hr delay, but they too failed to reproduce effects of dopaminergic medication on reinforcement learning. While supportive of a dopaminergic role in consolidation, this study failed to replicate previous findings on reinforcement learning.

Enhanced Experience Replay for Deep Reinforcement Learning

DTIC Science & Technology

2015-11-01

ARL-TR-7538 ● NOV 2015 US Army Research Laboratory Enhanced Experience Replay for Deep Reinforcement Learning by David Doria...Experience Replay for Deep Reinforcement Learning by David Doria, Bryan Dawson, and Manuel Vindiola Computational and Information Sciences Directorate...

Quantum-enhanced reinforcement learning for finite-episode games with discrete state spaces

NASA Astrophysics Data System (ADS)

Neukart, Florian; Von Dollen, David; Seidel, Christian; Compostella, Gabriele

2017-12-01

Quantum annealing algorithms belong to the class of metaheuristic tools, applicable for solving binary optimization problems. Hardware implementations of quantum annealing, such as the quantum annealing machines produced by D-Wave Systems, have been subject to multiple analyses in research, with the aim of characterizing the technology's usefulness for optimization and sampling tasks. Here, we present a way to partially embed both Monte Carlo policy iteration for finding an optimal policy on random observations, as well as how to embed n sub-optimal state-value functions for approximating an improved state-value function given a policy for finite horizon games with discrete state spaces on a D-Wave 2000Q quantum processing unit (QPU). We explain how both problems can be expressed as a quadratic unconstrained binary optimization (QUBO) problem, and show that quantum-enhanced Monte Carlo policy evaluation allows for finding equivalent or better state-value functions for a given policy with the same number episodes compared to a purely classical Monte Carlo algorithm. Additionally, we describe a quantum-classical policy learning algorithm. Our first and foremost aim is to explain how to represent and solve parts of these problems with the help of the QPU, and not to prove supremacy over every existing classical policy evaluation algorithm.

Risk-aware multi-armed bandit problem with application to portfolio selection

PubMed Central

Huo, Xiaoguang

2017-01-01

Sequential portfolio selection has attracted increasing interest in the machine learning and quantitative finance communities in recent years. As a mathematical framework for reinforcement learning policies, the stochastic multi-armed bandit problem addresses the primary difficulty in sequential decision-making under uncertainty, namely the exploration versus exploitation dilemma, and therefore provides a natural connection to portfolio selection. In this paper, we incorporate risk awareness into the classic multi-armed bandit setting and introduce an algorithm to construct portfolio. Through filtering assets based on the topological structure of the financial market and combining the optimal multi-armed bandit policy with the minimization of a coherent risk measure, we achieve a balance between risk and return. PMID:29291122

Risk-aware multi-armed bandit problem with application to portfolio selection.

PubMed

Huo, Xiaoguang; Fu, Feng

2017-11-01

Sequential portfolio selection has attracted increasing interest in the machine learning and quantitative finance communities in recent years. As a mathematical framework for reinforcement learning policies, the stochastic multi-armed bandit problem addresses the primary difficulty in sequential decision-making under uncertainty, namely the exploration versus exploitation dilemma, and therefore provides a natural connection to portfolio selection. In this paper, we incorporate risk awareness into the classic multi-armed bandit setting and introduce an algorithm to construct portfolio. Through filtering assets based on the topological structure of the financial market and combining the optimal multi-armed bandit policy with the minimization of a coherent risk measure, we achieve a balance between risk and return.

Prespeech motor learning in a neural network using reinforcement.

PubMed

Warlaumont, Anne S; Westermann, Gert; Buder, Eugene H; Oller, D Kimbrough

2013-02-01

Vocal motor development in infancy provides a crucial foundation for language development. Some significant early accomplishments include learning to control the process of phonation (the production of sound at the larynx) and learning to produce the sounds of one's language. Previous work has shown that social reinforcement shapes the kinds of vocalizations infants produce. We present a neural network model that provides an account of how vocal learning may be guided by reinforcement. The model consists of a self-organizing map that outputs to muscles of a realistic vocalization synthesizer. Vocalizations are spontaneously produced by the network. If a vocalization meets certain acoustic criteria, it is reinforced, and the weights are updated to make similar muscle activations increasingly likely to recur. We ran simulations of the model under various reinforcement criteria and tested the types of vocalizations it produced after learning in the different conditions. When reinforcement was contingent on the production of phonated (i.e. voiced) sounds, the network's post-learning productions were almost always phonated, whereas when reinforcement was not contingent on phonation, the network's post-learning productions were almost always not phonated. When reinforcement was contingent on both phonation and proximity to English vowels as opposed to Korean vowels, the model's post-learning productions were more likely to resemble the English vowels and vice versa. Copyright © 2012 Elsevier Ltd. All rights reserved.

Reconciling Reinforcement Learning Models with Behavioral Extinction and Renewal: Implications for Addiction, Relapse, and Problem Gambling

ERIC Educational Resources Information Center

Redish, A. David; Jensen, Steve; Johnson, Adam; Kurth-Nelson, Zeb

2007-01-01

Because learned associations are quickly renewed following extinction, the extinction process must include processes other than unlearning. However, reinforcement learning models, such as the temporal difference reinforcement learning (TDRL) model, treat extinction as an unlearning of associated value and are thus unable to capture renewal. TDRL…

Behavioral and neural properties of social reinforcement learning

PubMed Central

Jones, Rebecca M.; Somerville, Leah H.; Li, Jian; Ruberry, Erika J.; Libby, Victoria; Glover, Gary; Voss, Henning U.; Ballon, Douglas J.; Casey, BJ

2011-01-01

Social learning is critical for engaging in complex interactions with other individuals. Learning from positive social exchanges, such as acceptance from peers, may be similar to basic reinforcement learning. We formally test this hypothesis by developing a novel paradigm that is based upon work in non-human primates and human imaging studies of reinforcement learning. The probability of receiving positive social reinforcement from three distinct peers was parametrically manipulated while brain activity was recorded in healthy adults using event-related functional magnetic resonance imaging (fMRI). Over the course of the experiment, participants responded more quickly to faces of peers who provided more frequent positive social reinforcement, and rated them as more likeable. Modeling trial-by-trial learning showed ventral striatum and orbital frontal cortex activity correlated positively with forming expectations about receiving social reinforcement. Rostral anterior cingulate cortex activity tracked positively with modulations of expected value of the cues (peers). Together, the findings across three levels of analysis - social preferences, response latencies and modeling neural responses – are consistent with reinforcement learning theory and non-human primate electrophysiological studies of reward. This work highlights the fundamental influence of acceptance by one’s peers in altering subsequent behavior. PMID:21917787

The combination of appetitive and aversive reinforcers and the nature of their interaction during auditory learning.

PubMed

Ilango, A; Wetzel, W; Scheich, H; Ohl, F W

2010-03-31

Learned changes in behavior can be elicited by either appetitive or aversive reinforcers. It is, however, not clear whether the two types of motivation, (approaching appetitive stimuli and avoiding aversive stimuli) drive learning in the same or different ways, nor is their interaction understood in situations where the two types are combined in a single experiment. To investigate this question we have developed a novel learning paradigm for Mongolian gerbils, which not only allows rewards and punishments to be presented in isolation or in combination with each other, but also can use these opposite reinforcers to drive the same learned behavior. Specifically, we studied learning of tone-conditioned hurdle crossing in a shuttle box driven by either an appetitive reinforcer (brain stimulation reward) or an aversive reinforcer (electrical footshock), or by a combination of both. Combination of the two reinforcers potentiated speed of acquisition, led to maximum possible performance, and delayed extinction as compared to either reinforcer alone. Additional experiments, using partial reinforcement protocols and experiments in which one of the reinforcers was omitted after the animals had been previously trained with the combination of both reinforcers, indicated that appetitive and aversive reinforcers operated together but acted in different ways: in this particular experimental context, punishment appeared to be more effective for initial acquisition and reward more effective to maintain a high level of conditioned responses (CRs). The results imply that learning mechanisms in problem solving were maximally effective when the initial punishment of mistakes was combined with the subsequent rewarding of correct performance. Copyright 2010 IBRO. Published by Elsevier Ltd. All rights reserved.

Punishment Insensitivity and Impaired Reinforcement Learning in Preschoolers

ERIC Educational Resources Information Center

Briggs-Gowan, Margaret J.; Nichols, Sara R.; Voss, Joel; Zobel, Elvira; Carter, Alice S.; McCarthy, Kimberly J.; Pine, Daniel S.; Blair, James; Wakschlag, Lauren S.

2014-01-01

Background: Youth and adults with psychopathic traits display disrupted reinforcement learning. Advances in measurement now enable examination of this association in preschoolers. The current study examines relations between reinforcement learning in preschoolers and parent ratings of reduced responsiveness to socialization, conceptualized as a…

The cerebellum: a neural system for the study of reinforcement learning.

PubMed

Swain, Rodney A; Kerr, Abigail L; Thompson, Richard F

2011-01-01

In its strictest application, the term "reinforcement learning" refers to a computational approach to learning in which an agent (often a machine) interacts with a mutable environment to maximize reward through trial and error. The approach borrows essentials from several fields, most notably Computer Science, Behavioral Neuroscience, and Psychology. At the most basic level, a neural system capable of mediating reinforcement learning must be able to acquire sensory information about the external environment and internal milieu (either directly or through connectivities with other brain regions), must be able to select a behavior to be executed, and must be capable of providing evaluative feedback about the success of that behavior. Given that Psychology informs us that reinforcers, both positive and negative, are stimuli or consequences that increase the probability that the immediately antecedent behavior will be repeated and that reinforcer strength or viability is modulated by the organism's past experience with the reinforcer, its affect, and even the state of its muscles (e.g., eyes open or closed); it is the case that any neural system that supports reinforcement learning must also be sensitive to these same considerations. Once learning is established, such a neural system must finally be able to maintain continued response expression and prevent response drift. In this report, we examine both historical and recent evidence that the cerebellum satisfies all of these requirements. While we report evidence from a variety of learning paradigms, the majority of our discussion will focus on classical conditioning of the rabbit eye blink response as an ideal model system for the study of reinforcement and reinforcement learning.

Exact and Approximate Probabilistic Symbolic Execution

NASA Technical Reports Server (NTRS)

Luckow, Kasper; Pasareanu, Corina S.; Dwyer, Matthew B.; Filieri, Antonio; Visser, Willem

2014-01-01

Probabilistic software analysis seeks to quantify the likelihood of reaching a target event under uncertain environments. Recent approaches compute probabilities of execution paths using symbolic execution, but do not support nondeterminism. Nondeterminism arises naturally when no suitable probabilistic model can capture a program behavior, e.g., for multithreading or distributed systems. In this work, we propose a technique, based on symbolic execution, to synthesize schedulers that resolve nondeterminism to maximize the probability of reaching a target event. To scale to large systems, we also introduce approximate algorithms to search for good schedulers, speeding up established random sampling and reinforcement learning results through the quantification of path probabilities based on symbolic execution. We implemented the techniques in Symbolic PathFinder and evaluated them on nondeterministic Java programs. We show that our algorithms significantly improve upon a state-of- the-art statistical model checking algorithm, originally developed for Markov Decision Processes.

Neural Mechanism for Stochastic Behavior During a Competitive Game

PubMed Central

Soltani, Alireza; Lee, Daeyeol; Wang, Xiao-Jing

2006-01-01

Previous studies have shown that non-human primates can generate highly stochastic choice behavior, especially when this is required during a competitive interaction with another agent. To understand the neural mechanism of such dynamic choice behavior, we propose a biologically plausible model of decision making endowed with synaptic plasticity that follows a reward-dependent stochastic Hebbian learning rule. This model constitutes a biophysical implementation of reinforcement learning, and it reproduces salient features of behavioral data from an experiment with monkeys playing a matching pennies game. Due to interaction with an opponent and learning dynamics, the model generates quasi-random behavior robustly in spite of intrinsic biases. Furthermore, non-random choice behavior can also emerge when the model plays against a non-interactive opponent, as observed in the monkey experiment. Finally, when combined with a meta-learning algorithm, our model accounts for the slow drift in the animal’s strategy based on a process of reward maximization. PMID:17015181

Alterations in choice behavior by manipulations of world model.

PubMed

Green, C S; Benson, C; Kersten, D; Schrater, P

2010-09-14

How to compute initially unknown reward values makes up one of the key problems in reinforcement learning theory, with two basic approaches being used. Model-free algorithms rely on the accumulation of substantial amounts of experience to compute the value of actions, whereas in model-based learning, the agent seeks to learn the generative process for outcomes from which the value of actions can be predicted. Here we show that (i) "probability matching"-a consistent example of suboptimal choice behavior seen in humans-occurs in an optimal Bayesian model-based learner using a max decision rule that is initialized with ecologically plausible, but incorrect beliefs about the generative process for outcomes and (ii) human behavior can be strongly and predictably altered by the presence of cues suggestive of various generative processes, despite statistically identical outcome generation. These results suggest human decision making is rational and model based and not consistent with model-free learning.

Alterations in choice behavior by manipulations of world model

PubMed Central

Green, C. S.; Benson, C.; Kersten, D.; Schrater, P.

2010-01-01

How to compute initially unknown reward values makes up one of the key problems in reinforcement learning theory, with two basic approaches being used. Model-free algorithms rely on the accumulation of substantial amounts of experience to compute the value of actions, whereas in model-based learning, the agent seeks to learn the generative process for outcomes from which the value of actions can be predicted. Here we show that (i) “probability matching”—a consistent example of suboptimal choice behavior seen in humans—occurs in an optimal Bayesian model-based learner using a max decision rule that is initialized with ecologically plausible, but incorrect beliefs about the generative process for outcomes and (ii) human behavior can be strongly and predictably altered by the presence of cues suggestive of various generative processes, despite statistically identical outcome generation. These results suggest human decision making is rational and model based and not consistent with model-free learning. PMID:20805507

Neural Modularity Helps Organisms Evolve to Learn New Skills without Forgetting Old Skills

PubMed Central

Ellefsen, Kai Olav; Mouret, Jean-Baptiste; Clune, Jeff

2015-01-01

A long-standing goal in artificial intelligence is creating agents that can learn a variety of different skills for different problems. In the artificial intelligence subfield of neural networks, a barrier to that goal is that when agents learn a new skill they typically do so by losing previously acquired skills, a problem called catastrophic forgetting. That occurs because, to learn the new task, neural learning algorithms change connections that encode previously acquired skills. How networks are organized critically affects their learning dynamics. In this paper, we test whether catastrophic forgetting can be reduced by evolving modular neural networks. Modularity intuitively should reduce learning interference between tasks by separating functionality into physically distinct modules in which learning can be selectively turned on or off. Modularity can further improve learning by having a reinforcement learning module separate from sensory processing modules, allowing learning to happen only in response to a positive or negative reward. In this paper, learning takes place via neuromodulation, which allows agents to selectively change the rate of learning for each neural connection based on environmental stimuli (e.g. to alter learning in specific locations based on the task at hand). To produce modularity, we evolve neural networks with a cost for neural connections. We show that this connection cost technique causes modularity, confirming a previous result, and that such sparsely connected, modular networks have higher overall performance because they learn new skills faster while retaining old skills more and because they have a separate reinforcement learning module. Our results suggest (1) that encouraging modularity in neural networks may help us overcome the long-standing barrier of networks that cannot learn new skills without forgetting old ones, and (2) that one benefit of the modularity ubiquitous in the brains of natural animals might be to alleviate the problem of catastrophic forgetting. PMID:25837826

Neural modularity helps organisms evolve to learn new skills without forgetting old skills.

PubMed

Ellefsen, Kai Olav; Mouret, Jean-Baptiste; Clune, Jeff

2015-04-01

A long-standing goal in artificial intelligence is creating agents that can learn a variety of different skills for different problems. In the artificial intelligence subfield of neural networks, a barrier to that goal is that when agents learn a new skill they typically do so by losing previously acquired skills, a problem called catastrophic forgetting. That occurs because, to learn the new task, neural learning algorithms change connections that encode previously acquired skills. How networks are organized critically affects their learning dynamics. In this paper, we test whether catastrophic forgetting can be reduced by evolving modular neural networks. Modularity intuitively should reduce learning interference between tasks by separating functionality into physically distinct modules in which learning can be selectively turned on or off. Modularity can further improve learning by having a reinforcement learning module separate from sensory processing modules, allowing learning to happen only in response to a positive or negative reward. In this paper, learning takes place via neuromodulation, which allows agents to selectively change the rate of learning for each neural connection based on environmental stimuli (e.g. to alter learning in specific locations based on the task at hand). To produce modularity, we evolve neural networks with a cost for neural connections. We show that this connection cost technique causes modularity, confirming a previous result, and that such sparsely connected, modular networks have higher overall performance because they learn new skills faster while retaining old skills more and because they have a separate reinforcement learning module. Our results suggest (1) that encouraging modularity in neural networks may help us overcome the long-standing barrier of networks that cannot learn new skills without forgetting old ones, and (2) that one benefit of the modularity ubiquitous in the brains of natural animals might be to alleviate the problem of catastrophic forgetting.

Distinct prediction errors in mesostriatal circuits of the human brain mediate learning about the values of both states and actions: evidence from high-resolution fMRI

PubMed Central

Pauli, Wolfgang M.; Larsen, Tobias; Tyszka, J. Michael; O’Doherty, John P.

2017-01-01

Prediction-error signals consistent with formal models of “reinforcement learning” (RL) have repeatedly been found within dopaminergic nuclei of the midbrain and dopaminoceptive areas of the striatum. However, the precise form of the RL algorithms implemented in the human brain is not yet well determined. Here, we created a novel paradigm optimized to dissociate the subtypes of reward-prediction errors that function as the key computational signatures of two distinct classes of RL models—namely, “actor/critic” models and action-value-learning models (e.g., the Q-learning model). The state-value-prediction error (SVPE), which is independent of actions, is a hallmark of the actor/critic architecture, whereas the action-value-prediction error (AVPE) is the distinguishing feature of action-value-learning algorithms. To test for the presence of these prediction-error signals in the brain, we scanned human participants with a high-resolution functional magnetic-resonance imaging (fMRI) protocol optimized to enable measurement of neural activity in the dopaminergic midbrain as well as the striatal areas to which it projects. In keeping with the actor/critic model, the SVPE signal was detected in the substantia nigra. The SVPE was also clearly present in both the ventral striatum and the dorsal striatum. However, alongside these purely state-value-based computations we also found evidence for AVPE signals throughout the striatum. These high-resolution fMRI findings suggest that model-free aspects of reward learning in humans can be explained algorithmically with RL in terms of an actor/critic mechanism operating in parallel with a system for more direct action-value learning. PMID:29049406

Composition of web services using Markov decision processes and dynamic programming.

PubMed

Uc-Cetina, Víctor; Moo-Mena, Francisco; Hernandez-Ucan, Rafael

2015-01-01

We propose a Markov decision process model for solving the Web service composition (WSC) problem. Iterative policy evaluation, value iteration, and policy iteration algorithms are used to experimentally validate our approach, with artificial and real data. The experimental results show the reliability of the model and the methods employed, with policy iteration being the best one in terms of the minimum number of iterations needed to estimate an optimal policy, with the highest Quality of Service attributes. Our experimental work shows how the solution of a WSC problem involving a set of 100,000 individual Web services and where a valid composition requiring the selection of 1,000 services from the available set can be computed in the worst case in less than 200 seconds, using an Intel Core i5 computer with 6 GB RAM. Moreover, a real WSC problem involving only 7 individual Web services requires less than 0.08 seconds, using the same computational power. Finally, a comparison with two popular reinforcement learning algorithms, sarsa and Q-learning, shows that these algorithms require one or two orders of magnitude and more time than policy iteration, iterative policy evaluation, and value iteration to handle WSC problems of the same complexity.

Neural network based online simultaneous policy update algorithm for solving the HJI equation in nonlinear H∞ control.

PubMed

Wu, Huai-Ning; Luo, Biao

2012-12-01

It is well known that the nonlinear H∞ state feedback control problem relies on the solution of the Hamilton-Jacobi-Isaacs (HJI) equation, which is a nonlinear partial differential equation that has proven to be impossible to solve analytically. In this paper, a neural network (NN)-based online simultaneous policy update algorithm (SPUA) is developed to solve the HJI equation, in which knowledge of internal system dynamics is not required. First, we propose an online SPUA which can be viewed as a reinforcement learning technique for two players to learn their optimal actions in an unknown environment. The proposed online SPUA updates control and disturbance policies simultaneously; thus, only one iterative loop is needed. Second, the convergence of the online SPUA is established by proving that it is mathematically equivalent to Newton's method for finding a fixed point in a Banach space. Third, we develop an actor-critic structure for the implementation of the online SPUA, in which only one critic NN is needed for approximating the cost function, and a least-square method is given for estimating the NN weight parameters. Finally, simulation studies are provided to demonstrate the effectiveness of the proposed algorithm.

Variability in Dopamine Genes Dissociates Model-Based and Model-Free Reinforcement Learning.

PubMed

Doll, Bradley B; Bath, Kevin G; Daw, Nathaniel D; Frank, Michael J

2016-01-27

Considerable evidence suggests that multiple learning systems can drive behavior. Choice can proceed reflexively from previous actions and their associated outcomes, as captured by "model-free" learning algorithms, or flexibly from prospective consideration of outcomes that might occur, as captured by "model-based" learning algorithms. However, differential contributions of dopamine to these systems are poorly understood. Dopamine is widely thought to support model-free learning by modulating plasticity in striatum. Model-based learning may also be affected by these striatal effects, or by other dopaminergic effects elsewhere, notably on prefrontal working memory function. Indeed, prominent demonstrations linking striatal dopamine to putatively model-free learning did not rule out model-based effects, whereas other studies have reported dopaminergic modulation of verifiably model-based learning, but without distinguishing a prefrontal versus striatal locus. To clarify the relationships between dopamine, neural systems, and learning strategies, we combine a genetic association approach in humans with two well-studied reinforcement learning tasks: one isolating model-based from model-free behavior and the other sensitive to key aspects of striatal plasticity. Prefrontal function was indexed by a polymorphism in the COMT gene, differences of which reflect dopamine levels in the prefrontal cortex. This polymorphism has been associated with differences in prefrontal activity and working memory. Striatal function was indexed by a gene coding for DARPP-32, which is densely expressed in the striatum where it is necessary for synaptic plasticity. We found evidence for our hypothesis that variations in prefrontal dopamine relate to model-based learning, whereas variations in striatal dopamine function relate to model-free learning. Decisions can stem reflexively from their previously associated outcomes or flexibly from deliberative consideration of potential choice outcomes. Research implicates a dopamine-dependent striatal learning mechanism in the former type of choice. Although recent work has indicated that dopamine is also involved in flexible, goal-directed decision-making, it remains unclear whether it also contributes via striatum or via the dopamine-dependent working memory function of prefrontal cortex. We examined genetic indices of dopamine function in these regions and their relation to the two choice strategies. We found that striatal dopamine function related most clearly to the reflexive strategy, as previously shown, and that prefrontal dopamine related most clearly to the flexible strategy. These findings suggest that dissociable brain regions support dissociable choice strategies. Copyright © 2016 the authors 0270-6474/16/361211-12$15.00/0.

The role of first impression in operant learning.

PubMed

Shteingart, Hanan; Neiman, Tal; Loewenstein, Yonatan

2013-05-01

We quantified the effect of first experience on behavior in operant learning and studied its underlying computational principles. To that goal, we analyzed more than 200,000 choices in a repeated-choice experiment. We found that the outcome of the first experience has a substantial and lasting effect on participants' subsequent behavior, which we term outcome primacy. We found that this outcome primacy can account for much of the underweighting of rare events, where participants apparently underestimate small probabilities. We modeled behavior in this task using a standard, model-free reinforcement learning algorithm. In this model, the values of the different actions are learned over time and are used to determine the next action according to a predefined action-selection rule. We used a novel nonparametric method to characterize this action-selection rule and showed that the substantial effect of first experience on behavior is consistent with the reinforcement learning model if we assume that the outcome of first experience resets the values of the experienced actions, but not if we assume arbitrary initial conditions. Moreover, the predictive power of our resetting model outperforms previously published models regarding the aggregate choice behavior. These findings suggest that first experience has a disproportionately large effect on subsequent actions, similar to primacy effects in other fields of cognitive psychology. The mechanism of resetting of the initial conditions that underlies outcome primacy may thus also account for other forms of primacy. PsycINFO Database Record (c) 2013 APA, all rights reserved.

The role of GABAB receptors in human reinforcement learning.

PubMed

Ort, Andres; Kometer, Michael; Rohde, Judith; Seifritz, Erich; Vollenweider, Franz X

2014-10-01

Behavioral evidence from human studies suggests that the γ-aminobutyric acid type B receptor (GABAB receptor) agonist baclofen modulates reinforcement learning and reduces craving in patients with addiction spectrum disorders. However, in contrast to the well established role of dopamine in reinforcement learning, the mechanisms by which the GABAB receptor influences reinforcement learning in humans remain completely unknown. To further elucidate this issue, a cross-over, double-blind, placebo-controlled study was performed in healthy human subjects (N=15) to test the effects of baclofen (20 and 50mg p.o.) on probabilistic reinforcement learning. Outcomes were the feedback-induced P2 component of the event-related potential, the feedback-related negativity, and the P300 component of the event-related potential. Baclofen produced a reduction of P2 amplitude over the course of the experiment, but did not modulate the feedback-related negativity. Furthermore, there was a trend towards increased learning after baclofen administration relative to placebo over the course of the experiment. The present results extend previous theories of reinforcement learning, which focus on the importance of mesolimbic dopamine signaling, and indicate that stimulation of cortical GABAB receptors in a fronto-parietal network leads to better attentional allocation in reinforcement learning. This observation is a first step in our understanding of how baclofen may improve reinforcement learning in healthy subjects. Further studies with bigger sample sizes are needed to corroborate this conclusion and furthermore, test this effect in patients with addiction spectrum disorder. Copyright © 2014 Elsevier B.V. and ECNP. All rights reserved.

Effects of dopamine on reinforcement learning and consolidation in Parkinson’s disease

PubMed Central

Grogan, John P; Tsivos, Demitra; Smith, Laura; Knight, Brogan E; Bogacz, Rafal; Whone, Alan; Coulthard, Elizabeth J

2017-01-01

Emerging evidence suggests that dopamine may modulate learning and memory with important implications for understanding the neurobiology of memory and future therapeutic targeting. An influential hypothesis posits that dopamine biases reinforcement learning. More recent data also suggest an influence during both consolidation and retrieval. Eighteen Parkinson’s disease patients learned through feedback ON or OFF medication, with memory tested 24 hr later ON or OFF medication (4 conditions, within-subjects design with matched healthy control group). Patients OFF medication during learning decreased in memory accuracy over the following 24 hr. In contrast to previous studies, however, dopaminergic medication during learning and testing did not affect expression of positive or negative reinforcement. Two further experiments were run without the 24 hr delay, but they too failed to reproduce effects of dopaminergic medication on reinforcement learning. While supportive of a dopaminergic role in consolidation, this study failed to replicate previous findings on reinforcement learning. DOI: http://dx.doi.org/10.7554/eLife.26801.001 PMID:28691905

Fear of losing money? Aversive conditioning with secondary reinforcers.

PubMed

Delgado, M R; Labouliere, C D; Phelps, E A

2006-12-01

Money is a secondary reinforcer that acquires its value through social communication and interaction. In everyday human behavior and laboratory studies, money has been shown to influence appetitive or reward learning. It is unclear, however, if money has a similar impact on aversive learning. The goal of this study was to investigate the efficacy of money in aversive learning, comparing it with primary reinforcers that are traditionally used in fear conditioning paradigms. A series of experiments were conducted in which participants initially played a gambling game that led to a monetary gain. They were then presented with an aversive conditioning paradigm, with either shock (primary reinforcer) or loss of money (secondary reinforcer) as the unconditioned stimulus. Skin conductance responses and subjective ratings indicated that potential monetary loss modulated the conditioned response. Depending on the presentation context, the secondary reinforcer was as effective as the primary reinforcer during aversive conditioning. These results suggest that stimuli that acquire reinforcing properties through social communication and interaction, such as money, can effectively influence aversive learning.

Reinforcement learning and Tourette syndrome.

PubMed

Palminteri, Stefano; Pessiglione, Mathias

2013-01-01

In this chapter, we report the first experimental explorations of reinforcement learning in Tourette syndrome, realized by our team in the last few years. This report will be preceded by an introduction aimed to provide the reader with the state of the art of the knowledge concerning the neural bases of reinforcement learning at the moment of these studies and the scientific rationale beyond them. In short, reinforcement learning is learning by trial and error to maximize rewards and minimize punishments. This decision-making and learning process implicates the dopaminergic system projecting to the frontal cortex-basal ganglia circuits. A large body of evidence suggests that the dysfunction of the same neural systems is implicated in the pathophysiology of Tourette syndrome. Our results show that Tourette condition, as well as the most common pharmacological treatments (dopamine antagonists), affects reinforcement learning performance in these patients. Specifically, the results suggest a deficit in negative reinforcement learning, possibly underpinned by a functional hyperdopaminergia, which could explain the persistence of tics, despite their evident inadaptive (negative) value. This idea, together with the implications of these results in Tourette therapy and the future perspectives, is discussed in Section 4 of this chapter. © 2013 Elsevier Inc. All rights reserved.

Intelligence moderates reinforcement learning: a mini-review of the neural evidence

PubMed Central

2014-01-01

Our understanding of the neural basis of reinforcement learning and intelligence, two key factors contributing to human strivings, has progressed significantly recently. However, the overlap of these two lines of research, namely, how intelligence affects neural responses during reinforcement learning, remains uninvestigated. A mini-review of three existing studies suggests that higher IQ (especially fluid IQ) may enhance the neural signal of positive prediction error in dorsolateral prefrontal cortex, dorsal anterior cingulate cortex, and striatum, several brain substrates of reinforcement learning or intelligence. PMID:25185818

Intelligence moderates reinforcement learning: a mini-review of the neural evidence.

PubMed

Chen, Chong

2015-06-01

Our understanding of the neural basis of reinforcement learning and intelligence, two key factors contributing to human strivings, has progressed significantly recently. However, the overlap of these two lines of research, namely, how intelligence affects neural responses during reinforcement learning, remains uninvestigated. A mini-review of three existing studies suggests that higher IQ (especially fluid IQ) may enhance the neural signal of positive prediction error in dorsolateral prefrontal cortex, dorsal anterior cingulate cortex, and striatum, several brain substrates of reinforcement learning or intelligence. Copyright © 2015 the American Physiological Society.

Prespeech motor learning in a neural network using reinforcement☆

PubMed Central

Warlaumont, Anne S.; Westermann, Gert; Buder, Eugene H.; Oller, D. Kimbrough

2012-01-01

Vocal motor development in infancy provides a crucial foundation for language development. Some significant early accomplishments include learning to control the process of phonation (the production of sound at the larynx) and learning to produce the sounds of one’s language. Previous work has shown that social reinforcement shapes the kinds of vocalizations infants produce. We present a neural network model that provides an account of how vocal learning may be guided by reinforcement. The model consists of a self-organizing map that outputs to muscles of a realistic vocalization synthesizer. Vocalizations are spontaneously produced by the network. If a vocalization meets certain acoustic criteria, it is reinforced, and the weights are updated to make similar muscle activations increasingly likely to recur. We ran simulations of the model under various reinforcement criteria and tested the types of vocalizations it produced after learning in the differ-ent conditions. When reinforcement was contingent on the production of phonated (i.e. voiced) sounds, the network’s post learning productions were almost always phonated, whereas when reinforcement was not contingent on phonation, the network’s post-learning productions were almost always not phonated. When reinforcement was contingent on both phonation and proximity to English vowels as opposed to Korean vowels, the model’s post-learning productions were more likely to resemble the English vowels and vice versa. PMID:23275137

Reinforcement learning in complementarity game and population dynamics

NASA Astrophysics Data System (ADS)

Jost, Jürgen; Li, Wei

2014-02-01

We systematically test and compare different reinforcement learning schemes in a complementarity game [J. Jost and W. Li, Physica A 345, 245 (2005), 10.1016/j.physa.2004.07.005] played between members of two populations. More precisely, we study the Roth-Erev, Bush-Mosteller, and SoftMax reinforcement learning schemes. A modified version of Roth-Erev with a power exponent of 1.5, as opposed to 1 in the standard version, performs best. We also compare these reinforcement learning strategies with evolutionary schemes. This gives insight into aspects like the issue of quick adaptation as opposed to systematic exploration or the role of learning rates.

Online Learning of Genetic Network Programming and its Application to Prisoner’s Dilemma Game

NASA Astrophysics Data System (ADS)

Mabu, Shingo; Hirasawa, Kotaro; Hu, Jinglu; Murata, Junichi

A new evolutionary model with the network structure named Genetic Network Programming (GNP) has been proposed recently. GNP, that is, an expansion of GA and GP, represents solutions as a network structure and evolves it by using “offline learning (selection, mutation, crossover)”. GNP can memorize the past action sequences in the network flow, so it can deal with Partially Observable Markov Decision Process (POMDP) well. In this paper, in order to improve the ability of GNP, Q learning (an off-policy TD control algorithm) that is one of the famous online methods is introduced for online learning of GNP. Q learning is suitable for GNP because (1) in reinforcement learning, the rewards an agent will get in the future can be estimated, (2) TD control doesn’t need much memory and can learn quickly, and (3) off-policy is suitable in order to search for an optimal solution independently of the policy. Finally, in the simulations, online learning of GNP is applied to a player for “Prisoner’s dilemma game” and its ability for online adaptation is confirmed.

Learning, epigenetics, and computation: An extension on Fitch's proposal. Comment on “Toward a computational framework for cognitive biology: Unifying approaches from cognitive neuroscience and comparative cognition” by W. Tecumseh Fitch

NASA Astrophysics Data System (ADS)

Okanoya, Kazuo

2014-09-01

The comparative computational approach of Fitch [1] attempts to renew the classical David Marr paradigm of computation, algorithm, and implementation, by introducing evolutionary view of the relationship between neural architecture and cognition. This comparative evolutionary view provides constraints useful in narrowing down the problem space for both cognition and neural mechanisms. I will provide two examples from our own studies that reinforce and extend Fitch's proposal.

A new proof of the generalized Hamiltonian–Real calculus

PubMed Central

Gao, Hua; Mandic, Danilo P.

2016-01-01

The recently introduced generalized Hamiltonian–Real (GHR) calculus comprises, for the first time, the product and chain rules that makes it a powerful tool for quaternion-based optimization and adaptive signal processing. In this paper, we introduce novel dual relationships between the GHR calculus and multivariate real calculus, in order to provide a new, simpler proof of the GHR derivative rules. This further reinforces the theoretical foundation of the GHR calculus and provides a convenient methodology for generic extensions of real- and complex-valued learning algorithms to the quaternion domain.

Distributed Control with Collective Intelligence

NASA Technical Reports Server (NTRS)

Wolpert, David H.; Wheeler, Kevin R.; Tumer, Kagan

1998-01-01

We consider systems of interacting reinforcement learning (RL) algorithms that do not work at cross purposes , in that their collective behavior maximizes a global utility function. We call such systems COllective INtelligences (COINs). We present the theory of designing COINs. Then we present experiments validating that theory in the context of two distributed control problems: We show that COINs perform near-optimally in a difficult variant of Arthur's bar problem [Arthur] (and in particular avoid the tragedy of the commons for that problem), and we also illustrate optimal performance in the master-slave problem.

The prefrontal cortex and hybrid learning during iterative competitive games.

PubMed

Abe, Hiroshi; Seo, Hyojung; Lee, Daeyeol

2011-12-01

Behavioral changes driven by reinforcement and punishment are referred to as simple or model-free reinforcement learning. Animals can also change their behaviors by observing events that are neither appetitive nor aversive when these events provide new information about payoffs available from alternative actions. This is an example of model-based reinforcement learning and can be accomplished by incorporating hypothetical reward signals into the value functions for specific actions. Recent neuroimaging and single-neuron recording studies showed that the prefrontal cortex and the striatum are involved not only in reinforcement and punishment, but also in model-based reinforcement learning. We found evidence for both types of learning, and hence hybrid learning, in monkeys during simulated competitive games. In addition, in both the dorsolateral prefrontal cortex and orbitofrontal cortex, individual neurons heterogeneously encoded signals related to actual and hypothetical outcomes from specific actions, suggesting that both areas might contribute to hybrid learning. © 2011 New York Academy of Sciences.

Reinforcement Learning and Dopamine in Schizophrenia: Dimensions of Symptoms or Specific Features of a Disease Group?

PubMed Central

Deserno, Lorenz; Boehme, Rebecca; Heinz, Andreas; Schlagenhauf, Florian

2013-01-01

Abnormalities in reinforcement learning are a key finding in schizophrenia and have been proposed to be linked to elevated levels of dopamine neurotransmission. Behavioral deficits in reinforcement learning and their neural correlates may contribute to the formation of clinical characteristics of schizophrenia. The ability to form predictions about future outcomes is fundamental for environmental interactions and depends on neuronal teaching signals, like reward prediction errors. While aberrant prediction errors, that encode non-salient events as surprising, have been proposed to contribute to the formation of positive symptoms, a failure to build neural representations of decision values may result in negative symptoms. Here, we review behavioral and neuroimaging research in schizophrenia and focus on studies that implemented reinforcement learning models. In addition, we discuss studies that combined reinforcement learning with measures of dopamine. Thereby, we suggest how reinforcement learning abnormalities in schizophrenia may contribute to the formation of psychotic symptoms and may interact with cognitive deficits. These ideas point toward an interplay of more rigid versus flexible control over reinforcement learning. Pronounced deficits in the flexible or model-based domain may allow for a detailed characterization of well-established cognitive deficits in schizophrenia patients based on computational models of learning. Finally, we propose a framework based on the potentially crucial contribution of dopamine to dysfunctional reinforcement learning on the level of neural networks. Future research may strongly benefit from computational modeling but also requires further methodological improvement for clinical group studies. These research tools may help to improve our understanding of disease-specific mechanisms and may help to identify clinically relevant subgroups of the heterogeneous entity schizophrenia. PMID:24391603

Generalization of value in reinforcement learning by humans.

PubMed

Wimmer, G Elliott; Daw, Nathaniel D; Shohamy, Daphna

2012-04-01

Research in decision-making has focused on the role of dopamine and its striatal targets in guiding choices via learned stimulus-reward or stimulus-response associations, behavior that is well described by reinforcement learning theories. However, basic reinforcement learning is relatively limited in scope and does not explain how learning about stimulus regularities or relations may guide decision-making. A candidate mechanism for this type of learning comes from the domain of memory, which has highlighted a role for the hippocampus in learning of stimulus-stimulus relations, typically dissociated from the role of the striatum in stimulus-response learning. Here, we used functional magnetic resonance imaging and computational model-based analyses to examine the joint contributions of these mechanisms to reinforcement learning. Humans performed a reinforcement learning task with added relational structure, modeled after tasks used to isolate hippocampal contributions to memory. On each trial participants chose one of four options, but the reward probabilities for pairs of options were correlated across trials. This (uninstructed) relationship between pairs of options potentially enabled an observer to learn about option values based on experience with the other options and to generalize across them. We observed blood oxygen level-dependent (BOLD) activity related to learning in the striatum and also in the hippocampus. By comparing a basic reinforcement learning model to one augmented to allow feedback to generalize between correlated options, we tested whether choice behavior and BOLD activity were influenced by the opportunity to generalize across correlated options. Although such generalization goes beyond standard computational accounts of reinforcement learning and striatal BOLD, both choices and striatal BOLD activity were better explained by the augmented model. Consistent with the hypothesized role for the hippocampus in this generalization, functional connectivity between the ventral striatum and hippocampus was modulated, across participants, by the ability of the augmented model to capture participants' choice. Our results thus point toward an interactive model in which striatal reinforcement learning systems may employ relational representations typically associated with the hippocampus. © 2012 The Authors. European Journal of Neuroscience © 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd.

Instrumental learning and relearning in individuals with psychopathy and in patients with lesions involving the amygdala or orbitofrontal cortex.

PubMed

Mitchell, D G V; Fine, C; Richell, R A; Newman, C; Lumsden, J; Blair, K S; Blair, R J R

2006-05-01

Previous work has shown that individuals with psychopathy are impaired on some forms of associative learning, particularly stimulus-reinforcement learning (Blair et al., 2004; Newman & Kosson, 1986). Animal work suggests that the acquisition of stimulus-reinforcement associations requires the amygdala (Baxter & Murray, 2002). Individuals with psychopathy also show impoverished reversal learning (Mitchell, Colledge, Leonard, & Blair, 2002). Reversal learning is supported by the ventrolateral and orbitofrontal cortex (Rolls, 2004). In this paper we present experiments investigating stimulus-reinforcement learning and relearning in patients with lesions of the orbitofrontal cortex or amygdala, and individuals with developmental psychopathy without known trauma. The results are interpreted with reference to current neurocognitive models of stimulus-reinforcement learning, relearning, and developmental psychopathy. Copyright (c) 2006 APA, all rights reserved.

Model-based reinforcement learning with dimension reduction.

PubMed

Tangkaratt, Voot; Morimoto, Jun; Sugiyama, Masashi

2016-12-01

The goal of reinforcement learning is to learn an optimal policy which controls an agent to acquire the maximum cumulative reward. The model-based reinforcement learning approach learns a transition model of the environment from data, and then derives the optimal policy using the transition model. However, learning an accurate transition model in high-dimensional environments requires a large amount of data which is difficult to obtain. To overcome this difficulty, in this paper, we propose to combine model-based reinforcement learning with the recently developed least-squares conditional entropy (LSCE) method, which simultaneously performs transition model estimation and dimension reduction. We also further extend the proposed method to imitation learning scenarios. The experimental results show that policy search combined with LSCE performs well for high-dimensional control tasks including real humanoid robot control. Copyright © 2016 Elsevier Ltd. All rights reserved.

Reinforcement of Science Learning through Local Culture: A Delphi Study

ERIC Educational Resources Information Center

Nuangchalerm, Prasart

2008-01-01

This study aims to explore the ways to reinforce science learning through local culture by using Delphi technique. Twenty four participants in various fields of study were selected. The result of study provides a framework for reinforcement of science learning through local culture on the theme life and environment. (Contains 1 table.)

Autonomous Motion Learning for Intra-Vehicular Activity Space Robot

NASA Astrophysics Data System (ADS)

Watanabe, Yutaka; Yairi, Takehisa; Machida, Kazuo

Space robots will be needed in the future space missions. So far, many types of space robots have been developed, but in particular, Intra-Vehicular Activity (IVA) space robots that support human activities should be developed to reduce human-risks in space. In this paper, we study the motion learning method of an IVA space robot with the multi-link mechanism. The advantage point is that this space robot moves using reaction force of the multi-link mechanism and contact forces from the wall as space walking of an astronaut, not to use a propulsion. The control approach is determined based on a reinforcement learning with the actor-critic algorithm. We demonstrate to clear effectiveness of this approach using a 5-link space robot model by simulation. First, we simulate that a space robot learn the motion control including contact phase in two dimensional case. Next, we simulate that a space robot learn the motion control changing base attitude in three dimensional case.

Memristive device based learning for navigation in robots.

PubMed

Sarim, Mohammad; Kumar, Manish; Jha, Rashmi; Minai, Ali A

2017-11-08

Biomimetic robots have gained attention recently for various applications ranging from resource hunting to search and rescue operations during disasters. Biological species are known to intuitively learn from the environment, gather and process data, and make appropriate decisions. Such sophisticated computing capabilities in robots are difficult to achieve, especially if done in real-time with ultra-low energy consumption. Here, we present a novel memristive device based learning architecture for robots. Two terminal memristive devices with resistive switching of oxide layer are modeled in a crossbar array to develop a neuromorphic platform that can impart active real-time learning capabilities in a robot. This approach is validated by navigating a robot vehicle in an unknown environment with randomly placed obstacles. Further, the proposed scheme is compared with reinforcement learning based algorithms using local and global knowledge of the environment. The simulation as well as experimental results corroborate the validity and potential of the proposed learning scheme for robots. The results also show that our learning scheme approaches an optimal solution for some environment layouts in robot navigation.

Partial Planning Reinforcement Learning

DTIC Science & Technology

2012-08-31

Research Office P.O. Box 12211 Research Triangle Park, NC 27709-2211 15. SUBJECT TERMS Reinforcement Learning, Bayesian Optimization, Active ... Learning , Action Model Learning, Decision Theoretic Assistance Prasad Tadepalli, Alan Fern Oregon State University Office of Sponsored Programs Oregon State

Reinforcement learning in scheduling

NASA Technical Reports Server (NTRS)

Dietterich, Tom G.; Ok, Dokyeong; Zhang, Wei; Tadepalli, Prasad

1994-01-01

The goal of this research is to apply reinforcement learning methods to real-world problems like scheduling. In this preliminary paper, we show that learning to solve scheduling problems such as the Space Shuttle Payload Processing and the Automatic Guided Vehicle (AGV) scheduling can be usefully studied in the reinforcement learning framework. We discuss some of the special challenges posed by the scheduling domain to these methods and propose some possible solutions we plan to implement.

Neural Basis of Reinforcement Learning and Decision Making

PubMed Central

Lee, Daeyeol; Seo, Hyojung; Jung, Min Whan

2012-01-01

Reinforcement learning is an adaptive process in which an animal utilizes its previous experience to improve the outcomes of future choices. Computational theories of reinforcement learning play a central role in the newly emerging areas of neuroeconomics and decision neuroscience. In this framework, actions are chosen according to their value functions, which describe how much future reward is expected from each action. Value functions can be adjusted not only through reward and penalty, but also by the animal’s knowledge of its current environment. Studies have revealed that a large proportion of the brain is involved in representing and updating value functions and using them to choose an action. However, how the nature of a behavioral task affects the neural mechanisms of reinforcement learning remains incompletely understood. Future studies should uncover the principles by which different computational elements of reinforcement learning are dynamically coordinated across the entire brain. PMID:22462543

Effect of reinforcement learning on coordination of multiangent systems

NASA Astrophysics Data System (ADS)

Bukkapatnam, Satish T. S.; Gao, Greg

2000-12-01

For effective coordination of distributed environments involving multiagent systems, learning ability of each agent in the environment plays a crucial role. In this paper, we develop a simple group learning method based on reinforcement, and study its effect on coordination through application to a supply chain procurement scenario involving a computer manufacturer. Here, all parties are represented by self-interested, autonomous agents, each capable of performing specific simple tasks. They negotiate with each other to perform complex tasks and thus coordinate supply chain procurement. Reinforcement learning is intended to enable each agent to reach a best negotiable price within a shortest possible time. Our simulations of the application scenario under different learning strategies reveals the positive effects of reinforcement learning on an agent's as well as the system's performance.

The curse of planning: dissecting multiple reinforcement-learning systems by taxing the central executive.

PubMed

Otto, A Ross; Gershman, Samuel J; Markman, Arthur B; Daw, Nathaniel D

2013-05-01

A number of accounts of human and animal behavior posit the operation of parallel and competing valuation systems in the control of choice behavior. In these accounts, a flexible but computationally expensive model-based reinforcement-learning system has been contrasted with a less flexible but more efficient model-free reinforcement-learning system. The factors governing which system controls behavior-and under what circumstances-are still unclear. Following the hypothesis that model-based reinforcement learning requires cognitive resources, we demonstrated that having human decision makers perform a demanding secondary task engenders increased reliance on a model-free reinforcement-learning strategy. Further, we showed that, across trials, people negotiate the trade-off between the two systems dynamically as a function of concurrent executive-function demands, and people's choice latencies reflect the computational expenses of the strategy they employ. These results demonstrate that competition between multiple learning systems can be controlled on a trial-by-trial basis by modulating the availability of cognitive resources.

The Curse of Planning: Dissecting multiple reinforcement learning systems by taxing the central executive

PubMed Central

Otto, A. Ross; Gershman, Samuel J.; Markman, Arthur B.; Daw, Nathaniel D.

2013-01-01

A number of accounts of human and animal behavior posit the operation of parallel and competing valuation systems in the control of choice behavior. Along these lines, a flexible but computationally expensive model-based reinforcement learning system has been contrasted with a less flexible but more efficient model-free reinforcement learning system. The factors governing which system controls behavior—and under what circumstances—are still unclear. Based on the hypothesis that model-based reinforcement learning requires cognitive resources, we demonstrate that having human decision-makers perform a demanding secondary task engenders increased reliance on a model-free reinforcement learning strategy. Further, we show that across trials, people negotiate this tradeoff dynamically as a function of concurrent executive function demands and their choice latencies reflect the computational expenses of the strategy employed. These results demonstrate that competition between multiple learning systems can be controlled on a trial-by-trial basis by modulating the availability of cognitive resources. PMID:23558545

Efficient reinforcement learning of a reservoir network model of parametric working memory achieved with a cluster population winner-take-all readout mechanism.

PubMed

Cheng, Zhenbo; Deng, Zhidong; Hu, Xiaolin; Zhang, Bo; Yang, Tianming

2015-12-01

The brain often has to make decisions based on information stored in working memory, but the neural circuitry underlying working memory is not fully understood. Many theoretical efforts have been focused on modeling the persistent delay period activity in the prefrontal areas that is believed to represent working memory. Recent experiments reveal that the delay period activity in the prefrontal cortex is neither static nor homogeneous as previously assumed. Models based on reservoir networks have been proposed to model such a dynamical activity pattern. The connections between neurons within a reservoir are random and do not require explicit tuning. Information storage does not depend on the stable states of the network. However, it is not clear how the encoded information can be retrieved for decision making with a biologically realistic algorithm. We therefore built a reservoir-based neural network to model the neuronal responses of the prefrontal cortex in a somatosensory delayed discrimination task. We first illustrate that the neurons in the reservoir exhibit a heterogeneous and dynamical delay period activity observed in previous experiments. Then we show that a cluster population circuit decodes the information from the reservoir with a winner-take-all mechanism and contributes to the decision making. Finally, we show that the model achieves a good performance rapidly by shaping only the readout with reinforcement learning. Our model reproduces important features of previous behavior and neurophysiology data. We illustrate for the first time how task-specific information stored in a reservoir network can be retrieved with a biologically plausible reinforcement learning training scheme. Copyright © 2015 the American Physiological Society.

Computational Properties of the Hippocampus Increase the Efficiency of Goal-Directed Foraging through Hierarchical Reinforcement Learning

PubMed Central

Chalmers, Eric; Luczak, Artur; Gruber, Aaron J.

2016-01-01

The mammalian brain is thought to use a version of Model-based Reinforcement Learning (MBRL) to guide “goal-directed” behavior, wherein animals consider goals and make plans to acquire desired outcomes. However, conventional MBRL algorithms do not fully explain animals' ability to rapidly adapt to environmental changes, or learn multiple complex tasks. They also require extensive computation, suggesting that goal-directed behavior is cognitively expensive. We propose here that key features of processing in the hippocampus support a flexible MBRL mechanism for spatial navigation that is computationally efficient and can adapt quickly to change. We investigate this idea by implementing a computational MBRL framework that incorporates features inspired by computational properties of the hippocampus: a hierarchical representation of space, “forward sweeps” through future spatial trajectories, and context-driven remapping of place cells. We find that a hierarchical abstraction of space greatly reduces the computational load (mental effort) required for adaptation to changing environmental conditions, and allows efficient scaling to large problems. It also allows abstract knowledge gained at high levels to guide adaptation to new obstacles. Moreover, a context-driven remapping mechanism allows learning and memory of multiple tasks. Simulating dorsal or ventral hippocampal lesions in our computational framework qualitatively reproduces behavioral deficits observed in rodents with analogous lesions. The framework may thus embody key features of how the brain organizes model-based RL to efficiently solve navigation and other difficult tasks. PMID:28018203

B-tree search reinforcement learning for model based intelligent agent

NASA Astrophysics Data System (ADS)

Bhuvaneswari, S.; Vignashwaran, R.

2013-03-01

Agents trained by learning techniques provide a powerful approximation of active solutions for naive approaches. In this study using B - Trees implying reinforced learning the data search for information retrieval is moderated to achieve accuracy with minimum search time. The impact of variables and tactics applied in training are determined using reinforcement learning. Agents based on these techniques perform satisfactory baseline and act as finite agents based on the predetermined model against competitors from the course.

A common neural circuit mechanism for internally guided and externally reinforced forms of motor learning.

PubMed

Hisey, Erin; Kearney, Matthew Gene; Mooney, Richard

2018-04-01

The complex skills underlying verbal and musical expression can be learned without external punishment or reward, indicating their learning is internally guided. The neural mechanisms that mediate internally guided learning are poorly understood, but a circuit comprising dopamine-releasing neurons in the midbrain ventral tegmental area (VTA) and their targets in the basal ganglia are important to externally reinforced learning. Juvenile zebra finches copy a tutor song in a process that is internally guided and, in adulthood, can learn to modify the fundamental frequency (pitch) of a target syllable in response to external reinforcement with white noise. Here we combined intersectional genetic ablation of VTA neurons, reversible blockade of dopamine receptors in the basal ganglia, and singing-triggered optogenetic stimulation of VTA terminals to establish that a common VTA-basal ganglia circuit enables internally guided song copying and externally reinforced syllable pitch learning.

Dynamic route and departure time choice model based on self-adaptive reference point and reinforcement learning

NASA Astrophysics Data System (ADS)

Li, Xue-yan; Li, Xue-mei; Yang, Lingrun; Li, Jing

2018-07-01

Most of the previous studies on dynamic traffic assignment are based on traditional analytical framework, for instance, the idea of Dynamic User Equilibrium has been widely used in depicting both the route choice and the departure time choice. However, some recent studies have demonstrated that the dynamic traffic flow assignment largely depends on travelers' rationality degree, travelers' heterogeneity and what the traffic information the travelers have. In this paper, we develop a new self-adaptive multi agent model to depict travelers' behavior in Dynamic Traffic Assignment. We use Cumulative Prospect Theory with heterogeneous reference points to illustrate travelers' bounded rationality. We use reinforcement-learning model to depict travelers' route and departure time choosing behavior under the condition of imperfect information. We design the evolution rule of travelers' expected arrival time and the algorithm of traffic flow assignment. Compared with the traditional model, the self-adaptive multi agent model we proposed in this paper can effectively help travelers avoid the rush hour. Finally, we report and analyze the effect of travelers' group behavior on the transportation system, and give some insights into the relation between travelers' group behavior and the performance of transportation system.

Reinforcement-Learning-Based Robust Controller Design for Continuous-Time Uncertain Nonlinear Systems Subject to Input Constraints.

PubMed

Liu, Derong; Yang, Xiong; Wang, Ding; Wei, Qinglai

2015-07-01

The design of stabilizing controller for uncertain nonlinear systems with control constraints is a challenging problem. The constrained-input coupled with the inability to identify accurately the uncertainties motivates the design of stabilizing controller based on reinforcement-learning (RL) methods. In this paper, a novel RL-based robust adaptive control algorithm is developed for a class of continuous-time uncertain nonlinear systems subject to input constraints. The robust control problem is converted to the constrained optimal control problem with appropriately selecting value functions for the nominal system. Distinct from typical action-critic dual networks employed in RL, only one critic neural network (NN) is constructed to derive the approximate optimal control. Meanwhile, unlike initial stabilizing control often indispensable in RL, there is no special requirement imposed on the initial control. By utilizing Lyapunov's direct method, the closed-loop optimal control system and the estimated weights of the critic NN are proved to be uniformly ultimately bounded. In addition, the derived approximate optimal control is verified to guarantee the uncertain nonlinear system to be stable in the sense of uniform ultimate boundedness. Two simulation examples are provided to illustrate the effectiveness and applicability of the present approach.

Off-policy reinforcement learning for H∞ control design.

PubMed

Luo, Biao; Wu, Huai-Ning; Huang, Tingwen

2015-01-01

The H∞ control design problem is considered for nonlinear systems with unknown internal system model. It is known that the nonlinear H∞ control problem can be transformed into solving the so-called Hamilton-Jacobi-Isaacs (HJI) equation, which is a nonlinear partial differential equation that is generally impossible to be solved analytically. Even worse, model-based approaches cannot be used for approximately solving HJI equation, when the accurate system model is unavailable or costly to obtain in practice. To overcome these difficulties, an off-policy reinforcement leaning (RL) method is introduced to learn the solution of HJI equation from real system data instead of mathematical system model, and its convergence is proved. In the off-policy RL method, the system data can be generated with arbitrary policies rather than the evaluating policy, which is extremely important and promising for practical systems. For implementation purpose, a neural network (NN)-based actor-critic structure is employed and a least-square NN weight update algorithm is derived based on the method of weighted residuals. Finally, the developed NN-based off-policy RL method is tested on a linear F16 aircraft plant, and further applied to a rotational/translational actuator system.

Knockout crickets for the study of learning and memory: Dopamine receptor Dop1 mediates aversive but not appetitive reinforcement in crickets.

PubMed

Awata, Hiroko; Watanabe, Takahito; Hamanaka, Yoshitaka; Mito, Taro; Noji, Sumihare; Mizunami, Makoto

2015-11-02

Elucidation of reinforcement mechanisms in associative learning is an important subject in neuroscience. In mammals, dopamine neurons are thought to play critical roles in mediating both appetitive and aversive reinforcement. Our pharmacological studies suggested that octopamine and dopamine neurons mediate reward and punishment, respectively, in crickets, but recent studies in fruit-flies concluded that dopamine neurons mediates both reward and punishment, via the type 1 dopamine receptor Dop1. To resolve the discrepancy between studies in different insect species, we produced Dop1 knockout crickets using the CRISPR/Cas9 system and found that they are defective in aversive learning with sodium chloride punishment but not appetitive learning with water or sucrose reward. The results suggest that dopamine and octopamine neurons mediate aversive and appetitive reinforcement, respectively, in crickets. We suggest unexpected diversity in neurotransmitters mediating appetitive reinforcement between crickets and fruit-flies, although the neurotransmitter mediating aversive reinforcement is conserved. This study demonstrates usefulness of the CRISPR/Cas9 system for producing knockout animals for the study of learning and memory.

Social Cognition as Reinforcement Learning: Feedback Modulates Emotion Inference.

PubMed

Zaki, Jamil; Kallman, Seth; Wimmer, G Elliott; Ochsner, Kevin; Shohamy, Daphna

2016-09-01

Neuroscientific studies of social cognition typically employ paradigms in which perceivers draw single-shot inferences about the internal states of strangers. Real-world social inference features much different parameters: People often encounter and learn about particular social targets (e.g., friends) over time and receive feedback about whether their inferences are correct or incorrect. Here, we examined this process and, more broadly, the intersection between social cognition and reinforcement learning. Perceivers were scanned using fMRI while repeatedly encountering three social targets who produced conflicting visual and verbal emotional cues. Perceivers guessed how targets felt and received feedback about whether they had guessed correctly. Visual cues reliably predicted one target's emotion, verbal cues predicted a second target's emotion, and neither reliably predicted the third target's emotion. Perceivers successfully used this information to update their judgments over time. Furthermore, trial-by-trial learning signals-estimated using two reinforcement learning models-tracked activity in ventral striatum and ventromedial pFC, structures associated with reinforcement learning, and regions associated with updating social impressions, including TPJ. These data suggest that learning about others' emotions, like other forms of feedback learning, relies on domain-general reinforcement mechanisms as well as domain-specific social information processing.

Composition of Web Services Using Markov Decision Processes and Dynamic Programming

PubMed Central

Uc-Cetina, Víctor; Moo-Mena, Francisco; Hernandez-Ucan, Rafael

2015-01-01

We propose a Markov decision process model for solving the Web service composition (WSC) problem. Iterative policy evaluation, value iteration, and policy iteration algorithms are used to experimentally validate our approach, with artificial and real data. The experimental results show the reliability of the model and the methods employed, with policy iteration being the best one in terms of the minimum number of iterations needed to estimate an optimal policy, with the highest Quality of Service attributes. Our experimental work shows how the solution of a WSC problem involving a set of 100,000 individual Web services and where a valid composition requiring the selection of 1,000 services from the available set can be computed in the worst case in less than 200 seconds, using an Intel Core i5 computer with 6 GB RAM. Moreover, a real WSC problem involving only 7 individual Web services requires less than 0.08 seconds, using the same computational power. Finally, a comparison with two popular reinforcement learning algorithms, sarsa and Q-learning, shows that these algorithms require one or two orders of magnitude and more time than policy iteration, iterative policy evaluation, and value iteration to handle WSC problems of the same complexity. PMID:25874247

Probabilistic Reinforcement Learning in Adults with Autism Spectrum Disorders

PubMed Central

Solomon, Marjorie; Smith, Anne C.; Frank, Michael J.; Ly, Stanford; Carter, Cameron S.

2017-01-01

Background Autism spectrum disorders (ASDs) can be conceptualized as disorders of learning, however there have been few experimental studies taking this perspective. Methods We examined the probabilistic reinforcement learning performance of 28 adults with ASDs and 30 typically developing adults on a task requiring learning relationships between three stimulus pairs consisting of Japanese characters with feedback that was valid with different probabilities (80%, 70%, and 60%). Both univariate and Bayesian state–space data analytic methods were employed. Hypotheses were based on the extant literature as well as on neurobiological and computational models of reinforcement learning. Results Both groups learned the task after training. However, there were group differences in early learning in the first task block where individuals with ASDs acquired the most frequently accurately reinforced stimulus pair (80%) comparably to typically developing individuals; exhibited poorer acquisition of the less frequently reinforced 70% pair as assessed by state–space learning curves; and outperformed typically developing individuals on the near chance (60%) pair. Individuals with ASDs also demonstrated deficits in using positive feedback to exploit rewarded choices. Conclusions Results support the contention that individuals with ASDs are slower learners. Based on neurobiology and on the results of computational modeling, one interpretation of this pattern of findings is that impairments are related to deficits in flexible updating of reinforcement history as mediated by the orbito-frontal cortex, with spared functioning of the basal ganglia. This hypothesis about the pathophysiology of learning in ASDs can be tested using functional magnetic resonance imaging. PMID:21425243

Adaptive Constrained Optimal Control Design for Data-Based Nonlinear Discrete-Time Systems With Critic-Only Structure.

PubMed

Luo, Biao; Liu, Derong; Wu, Huai-Ning

2018-06-01

Reinforcement learning has proved to be a powerful tool to solve optimal control problems over the past few years. However, the data-based constrained optimal control problem of nonaffine nonlinear discrete-time systems has rarely been studied yet. To solve this problem, an adaptive optimal control approach is developed by using the value iteration-based Q-learning (VIQL) with the critic-only structure. Most of the existing constrained control methods require the use of a certain performance index and only suit for linear or affine nonlinear systems, which is unreasonable in practice. To overcome this problem, the system transformation is first introduced with the general performance index. Then, the constrained optimal control problem is converted to an unconstrained optimal control problem. By introducing the action-state value function, i.e., Q-function, the VIQL algorithm is proposed to learn the optimal Q-function of the data-based unconstrained optimal control problem. The convergence results of the VIQL algorithm are established with an easy-to-realize initial condition . To implement the VIQL algorithm, the critic-only structure is developed, where only one neural network is required to approximate the Q-function. The converged Q-function obtained from the critic-only VIQL method is employed to design the adaptive constrained optimal controller based on the gradient descent scheme. Finally, the effectiveness of the developed adaptive control method is tested on three examples with computer simulation.

Homeostatic Agent for General Environment

NASA Astrophysics Data System (ADS)

Yoshida, Naoto

2018-03-01

One of the essential aspect in biological agents is dynamic stability. This aspect, called homeostasis, is widely discussed in ethology, neuroscience and during the early stages of artificial intelligence. Ashby's homeostats are general-purpose learning machines for stabilizing essential variables of the agent in the face of general environments. However, despite their generality, the original homeostats couldn't be scaled because they searched their parameters randomly. In this paper, first we re-define the objective of homeostats as the maximization of a multi-step survival probability from the view point of sequential decision theory and probabilistic theory. Then we show that this optimization problem can be treated by using reinforcement learning algorithms with special agent architectures and theoretically-derived intrinsic reward functions. Finally we empirically demonstrate that agents with our architecture automatically learn to survive in a given environment, including environments with visual stimuli. Our survival agents can learn to eat food, avoid poison and stabilize essential variables through theoretically-derived single intrinsic reward formulations.

Role of dopamine D2 receptors in human reinforcement learning.

PubMed

Eisenegger, Christoph; Naef, Michael; Linssen, Anke; Clark, Luke; Gandamaneni, Praveen K; Müller, Ulrich; Robbins, Trevor W

2014-09-01

Influential neurocomputational models emphasize dopamine (DA) as an electrophysiological and neurochemical correlate of reinforcement learning. However, evidence of a specific causal role of DA receptors in learning has been less forthcoming, especially in humans. Here we combine, in a between-subjects design, administration of a high dose of the selective DA D2/3-receptor antagonist sulpiride with genetic analysis of the DA D2 receptor in a behavioral study of reinforcement learning in a sample of 78 healthy male volunteers. In contrast to predictions of prevailing models emphasizing DA's pivotal role in learning via prediction errors, we found that sulpiride did not disrupt learning, but rather induced profound impairments in choice performance. The disruption was selective for stimuli indicating reward, whereas loss avoidance performance was unaffected. Effects were driven by volunteers with higher serum levels of the drug, and in those with genetically determined lower density of striatal DA D2 receptors. This is the clearest demonstration to date for a causal modulatory role of the DA D2 receptor in choice performance that might be distinct from learning. Our findings challenge current reward prediction error models of reinforcement learning, and suggest that classical animal models emphasizing a role of postsynaptic DA D2 receptors in motivational aspects of reinforcement learning may apply to humans as well.

Role of Dopamine D2 Receptors in Human Reinforcement Learning

PubMed Central

Eisenegger, Christoph; Naef, Michael; Linssen, Anke; Clark, Luke; Gandamaneni, Praveen K; Müller, Ulrich; Robbins, Trevor W

2014-01-01

Influential neurocomputational models emphasize dopamine (DA) as an electrophysiological and neurochemical correlate of reinforcement learning. However, evidence of a specific causal role of DA receptors in learning has been less forthcoming, especially in humans. Here we combine, in a between-subjects design, administration of a high dose of the selective DA D2/3-receptor antagonist sulpiride with genetic analysis of the DA D2 receptor in a behavioral study of reinforcement learning in a sample of 78 healthy male volunteers. In contrast to predictions of prevailing models emphasizing DA's pivotal role in learning via prediction errors, we found that sulpiride did not disrupt learning, but rather induced profound impairments in choice performance. The disruption was selective for stimuli indicating reward, whereas loss avoidance performance was unaffected. Effects were driven by volunteers with higher serum levels of the drug, and in those with genetically determined lower density of striatal DA D2 receptors. This is the clearest demonstration to date for a causal modulatory role of the DA D2 receptor in choice performance that might be distinct from learning. Our findings challenge current reward prediction error models of reinforcement learning, and suggest that classical animal models emphasizing a role of postsynaptic DA D2 receptors in motivational aspects of reinforcement learning may apply to humans as well. PMID:24713613

Risk-sensitive reinforcement learning.

PubMed

Shen, Yun; Tobia, Michael J; Sommer, Tobias; Obermayer, Klaus

2014-07-01

We derive a family of risk-sensitive reinforcement learning methods for agents, who face sequential decision-making tasks in uncertain environments. By applying a utility function to the temporal difference (TD) error, nonlinear transformations are effectively applied not only to the received rewards but also to the true transition probabilities of the underlying Markov decision process. When appropriate utility functions are chosen, the agents' behaviors express key features of human behavior as predicted by prospect theory (Kahneman & Tversky, 1979 ), for example, different risk preferences for gains and losses, as well as the shape of subjective probability curves. We derive a risk-sensitive Q-learning algorithm, which is necessary for modeling human behavior when transition probabilities are unknown, and prove its convergence. As a proof of principle for the applicability of the new framework, we apply it to quantify human behavior in a sequential investment task. We find that the risk-sensitive variant provides a significantly better fit to the behavioral data and that it leads to an interpretation of the subject's responses that is indeed consistent with prospect theory. The analysis of simultaneously measured fMRI signals shows a significant correlation of the risk-sensitive TD error with BOLD signal change in the ventral striatum. In addition we find a significant correlation of the risk-sensitive Q-values with neural activity in the striatum, cingulate cortex, and insula that is not present if standard Q-values are used.

Neuromuscular control of the point to point and oscillatory movements of a sagittal arm with the actor-critic reinforcement learning method.

PubMed

Golkhou, Vahid; Parnianpour, Mohamad; Lucas, Caro

2005-04-01

In this study, we have used a single link system with a pair of muscles that are excited with alpha and gamma signals to achieve both point to point and oscillatory movements with variable amplitude and frequency.The system is highly nonlinear in all its physical and physiological attributes. The major physiological characteristics of this system are simultaneous activation of a pair of nonlinear muscle-like-actuators for control purposes, existence of nonlinear spindle-like sensors and Golgi tendon organ-like sensor, actions of gravity and external loading. Transmission delays are included in the afferent and efferent neural paths to account for a more accurate representation of the reflex loops.A reinforcement learning method with an actor-critic (AC) architecture instead of middle and low level of central nervous system (CNS), is used to track a desired trajectory. The actor in this structure is a two layer feedforward neural network and the critic is a model of the cerebellum. The critic is trained by state-action-reward-state-action (SARSA) method. The critic will train the actor by supervisory learning based on the prior experiences. Simulation studies of oscillatory movements based on the proposed algorithm demonstrate excellent tracking capability and after 280 epochs the RMS error for position and velocity profiles were 0.02, 0.04 rad and rad/s, respectively.

Toward an autonomous brain machine interface: integrating sensorimotor reward modulation and reinforcement learning.

PubMed

Marsh, Brandi T; Tarigoppula, Venkata S Aditya; Chen, Chen; Francis, Joseph T

2015-05-13

For decades, neurophysiologists have worked on elucidating the function of the cortical sensorimotor control system from the standpoint of kinematics or dynamics. Recently, computational neuroscientists have developed models that can emulate changes seen in the primary motor cortex during learning. However, these simulations rely on the existence of a reward-like signal in the primary sensorimotor cortex. Reward modulation of the primary sensorimotor cortex has yet to be characterized at the level of neural units. Here we demonstrate that single units/multiunits and local field potentials in the primary motor (M1) cortex of nonhuman primates (Macaca radiata) are modulated by reward expectation during reaching movements and that this modulation is present even while subjects passively view cursor motions that are predictive of either reward or nonreward. After establishing this reward modulation, we set out to determine whether we could correctly classify rewarding versus nonrewarding trials, on a moment-to-moment basis. This reward information could then be used in collaboration with reinforcement learning principles toward an autonomous brain-machine interface. The autonomous brain-machine interface would use M1 for both decoding movement intention and extraction of reward expectation information as evaluative feedback, which would then update the decoding algorithm as necessary. In the work presented here, we show that this, in theory, is possible. Copyright © 2015 the authors 0270-6474/15/357374-14$15.00/0.

Reinforcement learning in computer vision

NASA Astrophysics Data System (ADS)

Bernstein, A. V.; Burnaev, E. V.

2018-04-01

Nowadays, machine learning has become one of the basic technologies used in solving various computer vision tasks such as feature detection, image segmentation, object recognition and tracking. In many applications, various complex systems such as robots are equipped with visual sensors from which they learn state of surrounding environment by solving corresponding computer vision tasks. Solutions of these tasks are used for making decisions about possible future actions. It is not surprising that when solving computer vision tasks we should take into account special aspects of their subsequent application in model-based predictive control. Reinforcement learning is one of modern machine learning technologies in which learning is carried out through interaction with the environment. In recent years, Reinforcement learning has been used both for solving such applied tasks as processing and analysis of visual information, and for solving specific computer vision problems such as filtering, extracting image features, localizing objects in scenes, and many others. The paper describes shortly the Reinforcement learning technology and its use for solving computer vision problems.

Off-Policy Actor-Critic Structure for Optimal Control of Unknown Systems With Disturbances.

PubMed

Song, Ruizhuo; Lewis, Frank L; Wei, Qinglai; Zhang, Huaguang

2016-05-01

An optimal control method is developed for unknown continuous-time systems with unknown disturbances in this paper. The integral reinforcement learning (IRL) algorithm is presented to obtain the iterative control. Off-policy learning is used to allow the dynamics to be completely unknown. Neural networks are used to construct critic and action networks. It is shown that if there are unknown disturbances, off-policy IRL may not converge or may be biased. For reducing the influence of unknown disturbances, a disturbances compensation controller is added. It is proven that the weight errors are uniformly ultimately bounded based on Lyapunov techniques. Convergence of the Hamiltonian function is also proven. The simulation study demonstrates the effectiveness of the proposed optimal control method for unknown systems with disturbances.

From Recurrent Choice to Skill Learning: A Reinforcement-Learning Model

ERIC Educational Resources Information Center

Fu, Wai-Tat; Anderson, John R.

2006-01-01

The authors propose a reinforcement-learning mechanism as a model for recurrent choice and extend it to account for skill learning. The model was inspired by recent research in neurophysiological studies of the basal ganglia and provides an integrated explanation of recurrent choice behavior and skill learning. The behavior includes effects of…

Approximate dynamic programming for optimal stationary control with control-dependent noise.

PubMed

Jiang, Yu; Jiang, Zhong-Ping

2011-12-01

This brief studies the stochastic optimal control problem via reinforcement learning and approximate/adaptive dynamic programming (ADP). A policy iteration algorithm is derived in the presence of both additive and multiplicative noise using Itô calculus. The expectation of the approximated cost matrix is guaranteed to converge to the solution of some algebraic Riccati equation that gives rise to the optimal cost value. Moreover, the covariance of the approximated cost matrix can be reduced by increasing the length of time interval between two consecutive iterations. Finally, a numerical example is given to illustrate the efficiency of the proposed ADP methodology.

Adolescent-specific patterns of behavior and neural activity during social reinforcement learning

PubMed Central

Jones, Rebecca M.; Somerville, Leah H.; Li, Jian; Ruberry, Erika J.; Powers, Alisa; Mehta, Natasha; Dyke, Jonathan; Casey, BJ

2014-01-01

Humans are sophisticated social beings. Social cues from others are exceptionally salient, particularly during adolescence. Understanding how adolescents interpret and learn from variable social signals can provide insight into the observed shift in social sensitivity during this period. The current study tested 120 participants between the ages of 8 and 25 years on a social reinforcement learning task where the probability of receiving positive social feedback was parametrically manipulated. Seventy-eight of these participants completed the task during fMRI scanning. Modeling trial-by-trial learning, children and adults showed higher positive learning rates than adolescents, suggesting that adolescents demonstrated less differentiation in their reaction times for peers who provided more positive feedback. Forming expectations about receiving positive social reinforcement correlated with neural activity within the medial prefrontal cortex and ventral striatum across age. Adolescents, unlike children and adults, showed greater insular activity during positive prediction error learning and increased activity in the supplementary motor cortex and the putamen when receiving positive social feedback regardless of the expected outcome, suggesting that peer approval may motivate adolescents towards action. While different amounts of positive social reinforcement enhanced learning in children and adults, all positive social reinforcement equally motivated adolescents. Together, these findings indicate that sensitivity to peer approval during adolescence goes beyond simple reinforcement theory accounts and suggests possible explanations for how peers may motivate adolescent behavior. PMID:24550063

Adolescent-specific patterns of behavior and neural activity during social reinforcement learning.

PubMed

Jones, Rebecca M; Somerville, Leah H; Li, Jian; Ruberry, Erika J; Powers, Alisa; Mehta, Natasha; Dyke, Jonathan; Casey, B J

2014-06-01

Humans are sophisticated social beings. Social cues from others are exceptionally salient, particularly during adolescence. Understanding how adolescents interpret and learn from variable social signals can provide insight into the observed shift in social sensitivity during this period. The present study tested 120 participants between the ages of 8 and 25 years on a social reinforcement learning task where the probability of receiving positive social feedback was parametrically manipulated. Seventy-eight of these participants completed the task during fMRI scanning. Modeling trial-by-trial learning, children and adults showed higher positive learning rates than did adolescents, suggesting that adolescents demonstrated less differentiation in their reaction times for peers who provided more positive feedback. Forming expectations about receiving positive social reinforcement correlated with neural activity within the medial prefrontal cortex and ventral striatum across age. Adolescents, unlike children and adults, showed greater insular activity during positive prediction error learning and increased activity in the supplementary motor cortex and the putamen when receiving positive social feedback regardless of the expected outcome, suggesting that peer approval may motivate adolescents toward action. While different amounts of positive social reinforcement enhanced learning in children and adults, all positive social reinforcement equally motivated adolescents. Together, these findings indicate that sensitivity to peer approval during adolescence goes beyond simple reinforcement theory accounts and suggest possible explanations for how peers may motivate adolescent behavior.

Stochastic Reinforcement Benefits Skill Acquisition

ERIC Educational Resources Information Center

Dayan, Eran; Averbeck, Bruno B.; Richmond, Barry J.; Cohen, Leonardo G.

2014-01-01

Learning complex skills is driven by reinforcement, which facilitates both online within-session gains and retention of the acquired skills. Yet, in ecologically relevant situations, skills are often acquired when mapping between actions and rewarding outcomes is unknown to the learning agent, resulting in reinforcement schedules of a stochastic…

Multi-agent Reinforcement Learning Model for Effective Action Selection

NASA Astrophysics Data System (ADS)

Youk, Sang Jo; Lee, Bong Keun

Reinforcement learning is a sub area of machine learning concerned with how an agent ought to take actions in an environment so as to maximize some notion of long-term reward. In the case of multi-agent, especially, which state space and action space gets very enormous in compared to single agent, so it needs to take most effective measure available select the action strategy for effective reinforcement learning. This paper proposes a multi-agent reinforcement learning model based on fuzzy inference system in order to improve learning collect speed and select an effective action in multi-agent. This paper verifies an effective action select strategy through evaluation tests based on Robocop Keep away which is one of useful test-beds for multi-agent. Our proposed model can apply to evaluate efficiency of the various intelligent multi-agents and also can apply to strategy and tactics of robot soccer system.

Lung Nodule Detection via Deep Reinforcement Learning.

PubMed

Ali, Issa; Hart, Gregory R; Gunabushanam, Gowthaman; Liang, Ying; Muhammad, Wazir; Nartowt, Bradley; Kane, Michael; Ma, Xiaomei; Deng, Jun

2018-01-01

Lung cancer is the most common cause of cancer-related death globally. As a preventive measure, the United States Preventive Services Task Force (USPSTF) recommends annual screening of high risk individuals with low-dose computed tomography (CT). The resulting volume of CT scans from millions of people will pose a significant challenge for radiologists to interpret. To fill this gap, computer-aided detection (CAD) algorithms may prove to be the most promising solution. A crucial first step in the analysis of lung cancer screening results using CAD is the detection of pulmonary nodules, which may represent early-stage lung cancer. The objective of this work is to develop and validate a reinforcement learning model based on deep artificial neural networks for early detection of lung nodules in thoracic CT images. Inspired by the AlphaGo system, our deep learning algorithm takes a raw CT image as input and views it as a collection of states, and output a classification of whether a nodule is present or not. The dataset used to train our model is the LIDC/IDRI database hosted by the lung nodule analysis (LUNA) challenge. In total, there are 888 CT scans with annotations based on agreement from at least three out of four radiologists. As a result, there are 590 individuals having one or more nodules, and 298 having none. Our training results yielded an overall accuracy of 99.1% [sensitivity 99.2%, specificity 99.1%, positive predictive value (PPV) 99.1%, negative predictive value (NPV) 99.2%]. In our test, the results yielded an overall accuracy of 64.4% (sensitivity 58.9%, specificity 55.3%, PPV 54.2%, and NPV 60.0%). These early results show promise in solving the major issue of false positives in CT screening of lung nodules, and may help to save unnecessary follow-up tests and expenditures.

Lung Nodule Detection via Deep Reinforcement Learning

PubMed Central

Ali, Issa; Hart, Gregory R.; Gunabushanam, Gowthaman; Liang, Ying; Muhammad, Wazir; Nartowt, Bradley; Kane, Michael; Ma, Xiaomei; Deng, Jun

2018-01-01

Lung cancer is the most common cause of cancer-related death globally. As a preventive measure, the United States Preventive Services Task Force (USPSTF) recommends annual screening of high risk individuals with low-dose computed tomography (CT). The resulting volume of CT scans from millions of people will pose a significant challenge for radiologists to interpret. To fill this gap, computer-aided detection (CAD) algorithms may prove to be the most promising solution. A crucial first step in the analysis of lung cancer screening results using CAD is the detection of pulmonary nodules, which may represent early-stage lung cancer. The objective of this work is to develop and validate a reinforcement learning model based on deep artificial neural networks for early detection of lung nodules in thoracic CT images. Inspired by the AlphaGo system, our deep learning algorithm takes a raw CT image as input and views it as a collection of states, and output a classification of whether a nodule is present or not. The dataset used to train our model is the LIDC/IDRI database hosted by the lung nodule analysis (LUNA) challenge. In total, there are 888 CT scans with annotations based on agreement from at least three out of four radiologists. As a result, there are 590 individuals having one or more nodules, and 298 having none. Our training results yielded an overall accuracy of 99.1% [sensitivity 99.2%, specificity 99.1%, positive predictive value (PPV) 99.1%, negative predictive value (NPV) 99.2%]. In our test, the results yielded an overall accuracy of 64.4% (sensitivity 58.9%, specificity 55.3%, PPV 54.2%, and NPV 60.0%). These early results show promise in solving the major issue of false positives in CT screening of lung nodules, and may help to save unnecessary follow-up tests and expenditures. PMID:29713615

Quantum-Enhanced Machine Learning

NASA Astrophysics Data System (ADS)

Dunjko, Vedran; Taylor, Jacob M.; Briegel, Hans J.

2016-09-01

The emerging field of quantum machine learning has the potential to substantially aid in the problems and scope of artificial intelligence. This is only enhanced by recent successes in the field of classical machine learning. In this work we propose an approach for the systematic treatment of machine learning, from the perspective of quantum information. Our approach is general and covers all three main branches of machine learning: supervised, unsupervised, and reinforcement learning. While quantum improvements in supervised and unsupervised learning have been reported, reinforcement learning has received much less attention. Within our approach, we tackle the problem of quantum enhancements in reinforcement learning as well, and propose a systematic scheme for providing improvements. As an example, we show that quadratic improvements in learning efficiency, and exponential improvements in performance over limited time periods, can be obtained for a broad class of learning problems.

A Universal 3D Voxel Descriptor for Solid-State Material Informatics with Deep Convolutional Neural Networks.

PubMed

Kajita, Seiji; Ohba, Nobuko; Jinnouchi, Ryosuke; Asahi, Ryoji

2017-12-05

Material informatics (MI) is a promising approach to liberate us from the time-consuming Edisonian (trial and error) process for material discoveries, driven by machine-learning algorithms. Several descriptors, which are encoded material features to feed computers, were proposed in the last few decades. Especially to solid systems, however, their insufficient representations of three dimensionality of field quantities such as electron distributions and local potentials have critically hindered broad and practical successes of the solid-state MI. We develop a simple, generic 3D voxel descriptor that compacts any field quantities, in such a suitable way to implement convolutional neural networks (CNNs). We examine the 3D voxel descriptor encoded from the electron distribution by a regression test with 680 oxides data. The present scheme outperforms other existing descriptors in the prediction of Hartree energies that are significantly relevant to the long-wavelength distribution of the valence electrons. The results indicate that this scheme can forecast any functionals of field quantities just by learning sufficient amount of data, if there is an explicit correlation between the target properties and field quantities. This 3D descriptor opens a way to import prominent CNNs-based algorithms of supervised, semi-supervised and reinforcement learnings into the solid-state MI.

The drift diffusion model as the choice rule in reinforcement learning.

PubMed

Pedersen, Mads Lund; Frank, Michael J; Biele, Guido

2017-08-01

Current reinforcement-learning models often assume simplified decision processes that do not fully reflect the dynamic complexities of choice processes. Conversely, sequential-sampling models of decision making account for both choice accuracy and response time, but assume that decisions are based on static decision values. To combine these two computational models of decision making and learning, we implemented reinforcement-learning models in which the drift diffusion model describes the choice process, thereby capturing both within- and across-trial dynamics. To exemplify the utility of this approach, we quantitatively fit data from a common reinforcement-learning paradigm using hierarchical Bayesian parameter estimation, and compared model variants to determine whether they could capture the effects of stimulant medication in adult patients with attention-deficit hyperactivity disorder (ADHD). The model with the best relative fit provided a good description of the learning process, choices, and response times. A parameter recovery experiment showed that the hierarchical Bayesian modeling approach enabled accurate estimation of the model parameters. The model approach described here, using simultaneous estimation of reinforcement-learning and drift diffusion model parameters, shows promise for revealing new insights into the cognitive and neural mechanisms of learning and decision making, as well as the alteration of such processes in clinical groups.

Can model-free reinforcement learning explain deontological moral judgments?

PubMed

Ayars, Alisabeth

2016-05-01

Dual-systems frameworks propose that moral judgments are derived from both an immediate emotional response, and controlled/rational cognition. Recently Cushman (2013) proposed a new dual-system theory based on model-free and model-based reinforcement learning. Model-free learning attaches values to actions based on their history of reward and punishment, and explains some deontological, non-utilitarian judgments. Model-based learning involves the construction of a causal model of the world and allows for far-sighted planning; this form of learning fits well with utilitarian considerations that seek to maximize certain kinds of outcomes. I present three concerns regarding the use of model-free reinforcement learning to explain deontological moral judgment. First, many actions that humans find aversive from model-free learning are not judged to be morally wrong. Moral judgment must require something in addition to model-free learning. Second, there is a dearth of evidence for central predictions of the reinforcement account-e.g., that people with different reinforcement histories will, all else equal, make different moral judgments. Finally, to account for the effect of intention within the framework requires certain assumptions which lack support. These challenges are reasonable foci for future empirical/theoretical work on the model-free/model-based framework. Copyright © 2016 Elsevier B.V. All rights reserved.

The drift diffusion model as the choice rule in reinforcement learning

PubMed Central

Frank, Michael J.

2017-01-01

Current reinforcement-learning models often assume simplified decision processes that do not fully reflect the dynamic complexities of choice processes. Conversely, sequential-sampling models of decision making account for both choice accuracy and response time, but assume that decisions are based on static decision values. To combine these two computational models of decision making and learning, we implemented reinforcement-learning models in which the drift diffusion model describes the choice process, thereby capturing both within- and across-trial dynamics. To exemplify the utility of this approach, we quantitatively fit data from a common reinforcement-learning paradigm using hierarchical Bayesian parameter estimation, and compared model variants to determine whether they could capture the effects of stimulant medication in adult patients with attention-deficit hyper-activity disorder (ADHD). The model with the best relative fit provided a good description of the learning process, choices, and response times. A parameter recovery experiment showed that the hierarchical Bayesian modeling approach enabled accurate estimation of the model parameters. The model approach described here, using simultaneous estimation of reinforcement-learning and drift diffusion model parameters, shows promise for revealing new insights into the cognitive and neural mechanisms of learning and decision making, as well as the alteration of such processes in clinical groups. PMID:27966103

General functioning predicts reward and punishment learning in schizophrenia.

PubMed

Somlai, Zsuzsanna; Moustafa, Ahmed A; Kéri, Szabolcs; Myers, Catherine E; Gluck, Mark A

2011-04-01

Previous studies investigating feedback-driven reinforcement learning in patients with schizophrenia have provided mixed results. In this study, we explored the clinical predictors of reward and punishment learning using a probabilistic classification learning task. Patients with schizophrenia (n=40) performed similarly to healthy controls (n=30) on the classification learning task. However, more severe negative and general symptoms were associated with lower reward-learning performance, whereas poorer general psychosocial functioning was correlated with both lower reward- and punishment-learning performances. Multiple linear regression analyses indicated that general psychosocial functioning was the only significant predictor of reinforcement learning performance when education, antipsychotic dose, and positive, negative and general symptoms were included in the analysis. These results suggest a close relationship between reinforcement learning and general psychosocial functioning in schizophrenia. Published by Elsevier B.V.

Operant conditioning of enhanced pain sensitivity by heat-pain titration.

PubMed

Becker, Susanne; Kleinböhl, Dieter; Klossika, Iris; Hölzl, Rupert

2008-11-15

Operant conditioning mechanisms have been demonstrated to be important in the development of chronic pain. Most experimental studies have investigated the operant modulation of verbal pain reports with extrinsic reinforcement, such as verbal reinforcement. Whether this reflects actual changes in the subjective experience of the nociceptive stimulus remained unclear. This study replicates and extends our previous demonstration that enhanced pain sensitivity to prolonged heat-pain stimulation could be learned in healthy participants through intrinsic reinforcement (contingent changes in nociceptive input) independent of verbal pain reports. In addition, we examine whether different magnitudes of reinforcement differentially enhance pain sensitivity using an operant heat-pain titration paradigm. It is based on the previously developed non-verbal behavioral discrimination task for the assessment of sensitization, which uses discriminative down- or up-regulation of stimulus temperatures in response to changes in subjective intensity. In operant heat-pain titration, this discriminative behavior and not verbal pain report was contingently reinforced or punished by acute decreases or increases in heat-pain intensity. The magnitude of reinforcement was varied between three groups: low (N1=13), medium (N2=11) and high reinforcement (N3=12). Continuous reinforcement was applied to acquire and train the operant behavior, followed by partial reinforcement to analyze the underlying learning mechanisms. Results demonstrated that sensitization to prolonged heat-pain stimulation was enhanced by operant learning within 1h. The extent of sensitization was directly dependent on the received magnitude of reinforcement. Thus, operant learning mechanisms based on intrinsic reinforcement may provide an explanation for the gradual development of sustained hypersensitivity during pain that is becoming chronic.

Adaptive Sensor Tuning for Seismic Event Detection in Environment with Electromagnetic Noise

NASA Astrophysics Data System (ADS)

Ziegler, Abra E.

The goal of this research is to detect possible microseismic events at a carbon sequestration site. Data recorded on a continuous downhole microseismic array in the Farnsworth Field, an oil field in Northern Texas that hosts an ongoing carbon capture, utilization, and storage project, were evaluated using machine learning and reinforcement learning techniques to determine their effectiveness at seismic event detection on a dataset with electromagnetic noise. The data were recorded from a passive vertical monitoring array consisting of 16 levels of 3-component 15 Hz geophones installed in the field and continuously recording since January 2014. Electromagnetic and other noise recorded on the array has significantly impacted the utility of the data and it was necessary to characterize and filter the noise in order to attempt event detection. Traditional detection methods using short-term average/long-term average (STA/LTA) algorithms were evaluated and determined to be ineffective because of changing noise levels. To improve the performance of event detection and automatically and dynamically detect seismic events using effective data processing parameters, an adaptive sensor tuning (AST) algorithm developed by Sandia National Laboratories was utilized. AST exploits neuro-dynamic programming (reinforcement learning) trained with historic event data to automatically self-tune and determine optimal detection parameter settings. The key metric that guides the AST algorithm is consistency of each sensor with its nearest neighbors: parameters are automatically adjusted on a per station basis to be more or less sensitive to produce consistent agreement of detections in its neighborhood. The effects that changes in neighborhood configuration have on signal detection were explored, as it was determined that neighborhood-based detections significantly reduce the number of both missed and false detections in ground-truthed data. The performance of the AST algorithm was quantitatively evaluated during a variety of noise conditions and seismic detections were identified using AST and compared to ancillary injection data. During a period of CO2 injection in a nearby well to the monitoring array, 82% of seismic events were accurately detected, 13% of events were missed, and 5% of detections were determined to be false. Additionally, seismic risk was evaluated from the stress field and faulting regime at FWU to determine the likelihood of pressure perturbations to trigger slip on previously mapped faults. Faults oriented NW-SE were identified as requiring the smallest pore pressure changes to trigger slip and faults oriented N-S will also potentially be reactivated although this is less likely.

Adaptive optimal training of animal behavior

NASA Astrophysics Data System (ADS)

Bak, Ji Hyun; Choi, Jung Yoon; Akrami, Athena; Witten, Ilana; Pillow, Jonathan

Neuroscience experiments often require training animals to perform tasks designed to elicit various sensory, cognitive, and motor behaviors. Training typically involves a series of gradual adjustments of stimulus conditions and rewards in order to bring about learning. However, training protocols are usually hand-designed, and often require weeks or months to achieve a desired level of task performance. Here we combine ideas from reinforcement learning and adaptive optimal experimental design to formulate methods for efficient training of animal behavior. Our work addresses two intriguing problems at once: first, it seeks to infer the learning rules underlying an animal's behavioral changes during training; second, it seeks to exploit these rules to select stimuli that will maximize the rate of learning toward a desired objective. We develop and test these methods using data collected from rats during training on a two-interval sensory discrimination task. We show that we can accurately infer the parameters of a learning algorithm that describes how the animal's internal model of the task evolves over the course of training. We also demonstrate by simulation that our method can provide a substantial speedup over standard training methods.

Place preference and vocal learning rely on distinct reinforcers in songbirds.

PubMed

Murdoch, Don; Chen, Ruidong; Goldberg, Jesse H

2018-04-30

In reinforcement learning (RL) agents are typically tasked with maximizing a single objective function such as reward. But it remains poorly understood how agents might pursue distinct objectives at once. In machines, multiobjective RL can be achieved by dividing a single agent into multiple sub-agents, each of which is shaped by agent-specific reinforcement, but it remains unknown if animals adopt this strategy. Here we use songbirds to test if navigation and singing, two behaviors with distinct objectives, can be differentially reinforced. We demonstrate that strobe flashes aversively condition place preference but not song syllables. Brief noise bursts aversively condition song syllables but positively reinforce place preference. Thus distinct behavior-generating systems, or agencies, within a single animal can be shaped by correspondingly distinct reinforcement signals. Our findings suggest that spatially segregated vocal circuits can solve a credit assignment problem associated with multiobjective learning.

Dopamine-Dependent Reinforcement of Motor Skill Learning: Evidence from Gilles de la Tourette Syndrome

ERIC Educational Resources Information Center

Palminteri, Stefano; Lebreton, Mael; Worbe, Yulia; Hartmann, Andreas; Lehericy, Stephane; Vidailhet, Marie; Grabli, David; Pessiglione, Mathias

2011-01-01

Reinforcement learning theory has been extensively used to understand the neural underpinnings of instrumental behaviour. A central assumption surrounds dopamine signalling reward prediction errors, so as to update action values and ensure better choices in the future. However, educators may share the intuitive idea that reinforcements not only…

A study on expertise of agents and its effects on cooperative Q-learning.

PubMed

Araabi, Babak Nadjar; Mastoureshgh, Sahar; Ahmadabadi, Majid Nili

2007-04-01

Cooperation in learning (CL) can be realized in a multiagent system, if agents are capable of learning from both their own experiments and other agents' knowledge and expertise. Extra resources are exploited into higher efficiency and faster learning in CL as compared to that of individual learning (IL). In the real world, however, implementation of CL is not a straightforward task, in part due to possible differences in area of expertise (AOE). In this paper, reinforcement-learning homogenous agents are considered in an environment with multiple goals or tasks. As a result, they become expert in different domains with different amounts of expertness. Each agent uses a one-step Q-learning algorithm and is capable of exchanging its Q-table with those of its teammates. Two crucial questions are addressed in this paper: "How the AOE of an agent can be extracted?" and "How agents can improve their performance in CL by knowing their AOEs?" An algorithm is developed to extract the AOE based on state transitions as a gold standard from a behavioral point of view. Moreover, it is discussed that the AOE can be implicitly obtained through agents' expertness in the state level. Three new methods for CL through the combination of Q-tables are developed and examined for overall performance after CL. The performances of developed methods are compared with that of IL, strategy sharing (SS), and weighted SS (WSS). Obtained results show the superior performance of AOE-based methods as compared to that of existing CL methods, which do not use the notion of AOE. These results are very encouraging in support of the idea that "cooperation based on the AOE" performs better than the general CL methods.

Overcoming Learned Helplessness in Community College Students.

ERIC Educational Resources Information Center

Roueche, John E.; Mink, Oscar G.

1982-01-01

Reviews research on the effects of repeated experiences of helplessness and on locus of control. Identifies conditions necessary for overcoming learned helplessness; i.e., the potential for learning to occur; consistent reinforcement; relevant, valued reinforcers; and favorable psychological situation. Recommends eight ways for teachers to…

Dopamine D2 Receptor Signaling in the Nucleus Accumbens Comprises a Metabolic-Cognitive Brain Interface Regulating Metabolic Components of Glucose Reinforcement.

PubMed

Michaelides, Michael; Miller, Michael L; DiNieri, Jennifer A; Gomez, Juan L; Schwartz, Elizabeth; Egervari, Gabor; Wang, Gene Jack; Mobbs, Charles V; Volkow, Nora D; Hurd, Yasmin L

2017-11-01

Appetitive drive is influenced by coordinated interactions between brain circuits that regulate reinforcement and homeostatic signals that control metabolism. Glucose modulates striatal dopamine (DA) and regulates appetitive drive and reinforcement learning. Striatal DA D2 receptors (D2Rs) also regulate reinforcement learning and are implicated in glucose-related metabolic disorders. Nevertheless, interactions between striatal D2R and peripheral glucose have not been previously described. Here we show that manipulations involving striatal D2R signaling coincide with perseverative and impulsive-like responding for sucrose, a disaccharide consisting of fructose and glucose. Fructose conveys orosensory (ie, taste) reinforcement but does not convey metabolic (ie, nutrient-derived) reinforcement. Glucose however conveys orosensory reinforcement but unlike fructose, it is a major metabolic energy source, underlies sustained reinforcement, and activates striatal circuitry. We found that mice with deletion of dopamine- and cAMP-regulated neuronal phosphoprotein (DARPP-32) exclusively in D2R-expressing cells exhibited preferential D2R changes in the nucleus accumbens (NAc), a striatal region that critically regulates sucrose reinforcement. These changes coincided with perseverative and impulsive-like responding for sucrose pellets and sustained reinforcement learning of glucose-paired flavors. These mice were also characterized by significant glucose intolerance (ie, impaired glucose utilization). Systemic glucose administration significantly attenuated sucrose operant responding and D2R activation or blockade in the NAc bidirectionally modulated blood glucose levels and glucose tolerance. Collectively, these results implicate NAc D2R in regulating both peripheral glucose levels and glucose-dependent reinforcement learning behaviors and highlight the notion that glucose metabolic impairments arising from disrupted NAc D2R signaling are involved in compulsive and perseverative feeding behaviors.

Action, outcome, and value: a dual-system framework for morality.

PubMed

Cushman, Fiery

2013-08-01

Dual-system approaches to psychology explain the fundamental properties of human judgment, decision making, and behavior across diverse domains. Yet, the appropriate characterization of each system is a source of debate. For instance, a large body of research on moral psychology makes use of the contrast between "emotional" and "rational/cognitive" processes, yet even the chief proponents of this division recognize its shortcomings. Largely independently, research in the computational neurosciences has identified a broad division between two algorithms for learning and choice derived from formal models of reinforcement learning. One assigns value to actions intrinsically based on past experience, while another derives representations of value from an internally represented causal model of the world. This division between action- and outcome-based value representation provides an ideal framework for a dual-system theory in the moral domain.

Deep reinforcement learning for automated radiation adaptation in lung cancer.

PubMed

Tseng, Huan-Hsin; Luo, Yi; Cui, Sunan; Chien, Jen-Tzung; Ten Haken, Randall K; Naqa, Issam El

2017-12-01

To investigate deep reinforcement learning (DRL) based on historical treatment plans for developing automated radiation adaptation protocols for nonsmall cell lung cancer (NSCLC) patients that aim to maximize tumor local control at reduced rates of radiation pneumonitis grade 2 (RP2). In a retrospective population of 114 NSCLC patients who received radiotherapy, a three-component neural networks framework was developed for deep reinforcement learning (DRL) of dose fractionation adaptation. Large-scale patient characteristics included clinical, genetic, and imaging radiomics features in addition to tumor and lung dosimetric variables. First, a generative adversarial network (GAN) was employed to learn patient population characteristics necessary for DRL training from a relatively limited sample size. Second, a radiotherapy artificial environment (RAE) was reconstructed by a deep neural network (DNN) utilizing both original and synthetic data (by GAN) to estimate the transition probabilities for adaptation of personalized radiotherapy patients' treatment courses. Third, a deep Q-network (DQN) was applied to the RAE for choosing the optimal dose in a response-adapted treatment setting. This multicomponent reinforcement learning approach was benchmarked against real clinical decisions that were applied in an adaptive dose escalation clinical protocol. In which, 34 patients were treated based on avid PET signal in the tumor and constrained by a 17.2% normal tissue complication probability (NTCP) limit for RP2. The uncomplicated cure probability (P+) was used as a baseline reward function in the DRL. Taking our adaptive dose escalation protocol as a blueprint for the proposed DRL (GAN + RAE + DQN) architecture, we obtained an automated dose adaptation estimate for use at ∼2/3 of the way into the radiotherapy treatment course. By letting the DQN component freely control the estimated adaptive dose per fraction (ranging from 1-5 Gy), the DRL automatically favored dose escalation/de-escalation between 1.5 and 3.8 Gy, a range similar to that used in the clinical protocol. The same DQN yielded two patterns of dose escalation for the 34 test patients, but with different reward variants. First, using the baseline P+ reward function, individual adaptive fraction doses of the DQN had similar tendencies to the clinical data with an RMSE = 0.76 Gy; but adaptations suggested by the DQN were generally lower in magnitude (less aggressive). Second, by adjusting the P+ reward function with higher emphasis on mitigating local failure, better matching of doses between the DQN and the clinical protocol was achieved with an RMSE = 0.5 Gy. Moreover, the decisions selected by the DQN seemed to have better concordance with patients eventual outcomes. In comparison, the traditional temporal difference (TD) algorithm for reinforcement learning yielded an RMSE = 3.3 Gy due to numerical instabilities and lack of sufficient learning. We demonstrated that automated dose adaptation by DRL is a feasible and a promising approach for achieving similar results to those chosen by clinicians. The process may require customization of the reward function if individual cases were to be considered. However, development of this framework into a fully credible autonomous system for clinical decision support would require further validation on larger multi-institutional datasets. © 2017 American Association of Physicists in Medicine.

Implementation of a Space Communications Cognitive Engine

NASA Technical Reports Server (NTRS)

Hackett, Timothy M.; Bilen, Sven G.; Ferreira, Paulo Victor R.; Wyglinski, Alexander M.; Reinhart, Richard C.

2017-01-01

Although communications-based cognitive engines have been proposed, very few have been implemented in a full system, especially in a space communications system. In this paper, we detail the implementation of a multi-objective reinforcement-learning algorithm and deep artificial neural networks for the use as a radio-resource-allocation controller. The modular software architecture presented encourages re-use and easy modification for trying different algorithms. Various trade studies involved with the system implementation and integration are discussed. These include the choice of software libraries that provide platform flexibility and promote reusability, choices regarding the deployment of this cognitive engine within a system architecture using the DVB-S2 standard and commercial hardware, and constraints placed on the cognitive engine caused by real-world radio constraints. The implemented radio-resource allocation-management controller was then integrated with the larger spaceground system developed by NASA Glenn Research Center (GRC).

Fuzzy logic in autonomous orbital operations

NASA Technical Reports Server (NTRS)

Lea, Robert N.; Jani, Yashvant

1991-01-01

Fuzzy logic can be used advantageously in autonomous orbital operations that require the capability of handling imprecise measurements from sensors. Several applications are underway to investigate fuzzy logic approaches and develop guidance and control algorithms for autonomous orbital operations. Translational as well as rotational control of a spacecraft have been demonstrated using space shuttle simulations. An approach to a camera tracking system has been developed to support proximity operations and traffic management around the Space Station Freedom. Pattern recognition and object identification algorithms currently under development will become part of this camera system at an appropriate level in the future. A concept to control environment and life support systems for large Lunar based crew quarters is also under development. Investigations in the area of reinforcement learning, utilizing neural networks, combined with a fuzzy logic controller, are planned as a joint project with the Ames Research Center.

The Effects of Partial Reinforcement in the Acquisition and Extinction of Recurrent Serial Patterns.

ERIC Educational Resources Information Center

Dockstader, Steven L.

The purpose of these 2 experiments was to determine whether sequential response pattern behavior is affected by partial reinforcement in the same way as other behavior systems. The first experiment investigated the partial reinforcement extinction effects (PREE) in a sequential concept learning task where subjects were required to learn a…

Microstimulation of the Human Substantia Nigra Alters Reinforcement Learning

PubMed Central

Ramayya, Ashwin G.; Misra, Amrit

2014-01-01

Animal studies have shown that substantia nigra (SN) dopaminergic (DA) neurons strengthen action–reward associations during reinforcement learning, but their role in human learning is not known. Here, we applied microstimulation in the SN of 11 patients undergoing deep brain stimulation surgery for the treatment of Parkinson's disease as they performed a two-alternative probability learning task in which rewards were contingent on stimuli, rather than actions. Subjects demonstrated decreased learning from reward trials that were accompanied by phasic SN microstimulation compared with reward trials without stimulation. Subjects who showed large decreases in learning also showed an increased bias toward repeating actions after stimulation trials; therefore, stimulation may have decreased learning by strengthening action–reward associations rather than stimulus–reward associations. Our findings build on previous studies implicating SN DA neurons in preferentially strengthening action–reward associations during reinforcement learning. PMID:24828643

An Investigation of Ways to Reduce the Failure Rate of Student Pilots during Flying Training in the Royal Australian Air Force.

DTIC Science & Technology

1987-09-01

Luthans (28) expanded the concept of learning as follows: 1. Learning involves a change, though not necessarily an improvement, in behaviour. Learning...that results in an unpleasant outcome is not likely to be repeated (36:244). Luthans and Kreitner (27) described the various forms of reinforcement as...four 33 alternatives (defined previously on page 24 and taken from Luthans ) of positive reinforcement, negative reinforcement, extinction and punishment

Mastery Learning through Individualized Instruction: A Reinforcement Strategy

ERIC Educational Resources Information Center

Sagy, John; Ravi, R.; Ananthasayanam, R.

2009-01-01

The present study attempts to gauge the effect of individualized instructional methods as a reinforcement strategy for mastery learning. Among various individualized instructional methods, the study focuses on PIM (Programmed Instructional Method) and CAIM (Computer Assisted Instruction Method). Mastery learning is a process where students achieve…

Working Memory and Reinforcement Schedule Jointly Determine Reinforcement Learning in Children: Potential Implications for Behavioral Parent Training

PubMed Central

Segers, Elien; Beckers, Tom; Geurts, Hilde; Claes, Laurence; Danckaerts, Marina; van der Oord, Saskia

2018-01-01

Introduction: Behavioral Parent Training (BPT) is often provided for childhood psychiatric disorders. These disorders have been shown to be associated with working memory impairments. BPT is based on operant learning principles, yet how operant principles shape behavior (through the partial reinforcement (PRF) extinction effect, i.e., greater resistance to extinction that is created when behavior is reinforced partially rather than continuously) and the potential role of working memory therein is scarcely studied in children. This study explored the PRF extinction effect and the role of working memory therein using experimental tasks in typically developing children. Methods: Ninety-seven children (age 6–10) completed a working memory task and an operant learning task, in which children acquired a response-sequence rule under either continuous or PRF (120 trials), followed by an extinction phase (80 trials). Data of 88 children were used for analysis. Results: The PRF extinction effect was confirmed: We observed slower acquisition and extinction in the PRF condition as compared to the continuous reinforcement (CRF) condition. Working memory was negatively related to acquisition but not extinction performance. Conclusion: Both reinforcement contingencies and working memory relate to acquisition performance. Potential implications for BPT are that decreasing working memory load may enhance the chance of optimally learning through reinforcement. PMID:29643822

A novel microaneurysms detection approach based on convolutional neural networks with reinforcement sample learning algorithm.

PubMed

Budak, Umit; Şengür, Abdulkadir; Guo, Yanhui; Akbulut, Yaman

2017-12-01

Microaneurysms (MAs) are known as early signs of diabetic-retinopathy which are called red lesions in color fundus images. Detection of MAs in fundus images needs highly skilled physicians or eye angiography. Eye angiography is an invasive and expensive procedure. Therefore, an automatic detection system to identify the MAs locations in fundus images is in demand. In this paper, we proposed a system to detect the MAs in colored fundus images. The proposed method composed of three stages. In the first stage, a series of pre-processing steps are used to make the input images more convenient for MAs detection. To this end, green channel decomposition, Gaussian filtering, median filtering, back ground determination, and subtraction operations are applied to input colored fundus images. After pre-processing, a candidate MAs extraction procedure is applied to detect potential regions. A five-stepped procedure is adopted to get the potential MA locations. Finally, deep convolutional neural network (DCNN) with reinforcement sample learning strategy is used to train the proposed system. The DCNN is trained with color image patches which are collected from ground-truth MA locations and non-MA locations. We conducted extensive experiments on ROC dataset to evaluate of our proposal. The results are encouraging.

Design and implementation of a random neural network routing engine.

PubMed

Kocak, T; Seeber, J; Terzioglu, H

2003-01-01

Random neural network (RNN) is an analytically tractable spiked neural network model that has been implemented in software for a wide range of applications for over a decade. This paper presents the hardware implementation of the RNN model. Recently, cognitive packet networks (CPN) is proposed as an alternative packet network architecture where there is no routing table, instead the RNN based reinforcement learning is used to route packets. Particularly, we describe implementation details for the RNN based routing engine of a CPN network processor chip: the smart packet processor (SPP). The SPP is a dual port device that stores, modifies, and interprets the defining characteristics of multiple RNN models. In addition to hardware design improvements over the software implementation such as the dual access memory, output calculation step, and reduced output calculation module, this paper introduces a major modification to the reinforcement learning algorithm used in the original CPN specification such that the number of weight terms are reduced from 2n/sup 2/ to 2n. This not only yields significant memory savings, but it also simplifies the calculations for the steady state probabilities (neuron outputs in RNN). Simulations have been conducted to confirm the proper functionality for the isolated SPP design as well as for the multiple SPP's in a networked environment.

Time-Extended Payoffs for Collectives of Autonomous Agents

NASA Technical Reports Server (NTRS)

Tumer, Kagan; Agogino, Adrian K.

2002-01-01

A collective is a set of self-interested agents which try to maximize their own utilities, along with a a well-defined, time-extended world utility function which rates the performance of the entire system. In this paper, we use theory of collectives to design time-extended payoff utilities for agents that are both aligned with the world utility, and are "learnable", i.e., the agents can readily see how their behavior affects their utility. We show that in systems where each agent aims to optimize such payoff functions, coordination arises as a byproduct of the agents selfishly pursuing their own goals. A game theoretic analysis shows that such payoff functions have the net effect of aligning the Nash equilibrium, Pareto optimal solution and world utility optimum, thus eliminating undesirable behavior such as agents working at cross-purposes. We then apply collective-based payoff functions to the token collection in a gridworld problem where agents need to optimize the aggregate value of tokens collected across an episode of finite duration (i.e., an abstracted version of rovers on Mars collecting scientifically interesting rock samples, subject to power limitations). We show that, regardless of the initial token distribution, reinforcement learning agents using collective-based payoff functions significantly outperform both natural extensions of single agent algorithms and global reinforcement learning solutions based on "team games".

Instructional control of reinforcement learning: A behavioral and neurocomputational investigation

PubMed Central

Doll, Bradley B.; Jacobs, W. Jake; Sanfey, Alan G.; Frank, Michael J.

2011-01-01

Humans learn how to behave directly through environmental experience and indirectly through rules and instructions. Behavior analytic research has shown that instructions can control behavior, even when such behavior leads to sub-optimal outcomes (Hayes, S. (Ed.). 1989. Rule-governed behavior: cognition, contingencies, and instructional control. Plenum Press.). Here we examine the control of behavior through instructions in a reinforcement learning task known to depend on striatal dopaminergic function. Participants selected between probabilistically reinforced stimuli, and were (incorrectly) told that a specific stimulus had the highest (or lowest) reinforcement probability. Despite experience to the contrary, instructions drove choice behavior. We present neural network simulations that capture the interactions between instruction-driven and reinforcement-driven behavior via two potential neural circuits: one in which the striatum is inaccurately trained by instruction representations coming from prefrontal cortex/hippocampus (PFC/HC), and another in which the striatum learns the environmentally based reinforcement contingencies, but is “overridden” at decision output. Both models capture the core behavioral phenomena but, because they differ fundamentally on what is learned, make distinct predictions for subsequent behavioral and neuroimaging experiments. Finally, we attempt to distinguish between the proposed computational mechanisms governing instructed behavior by fitting a series of abstract “Q-learning” and Bayesian models to subject data. The best-fitting model supports one of the neural models, suggesting the existence of a “confirmation bias” in which the PFC/HC system trains the reinforcement system by amplifying outcomes that are consistent with instructions while diminishing inconsistent outcomes. PMID:19595993

How partial reinforcement of food cues affects the extinction and reacquisition of appetitive responses. A new model for dieting success?

PubMed

van den Akker, Karolien; Havermans, Remco C; Bouton, Mark E; Jansen, Anita

2014-10-01

Animals and humans can easily learn to associate an initially neutral cue with food intake through classical conditioning, but extinction of learned appetitive responses can be more difficult. Intermittent or partial reinforcement of food cues causes especially persistent behaviour in animals: after exposure to such learning schedules, the decline in responding that occurs during extinction is slow. After extinction, increases in responding with renewed reinforcement of food cues (reacquisition) might be less rapid after acquisition with partial reinforcement. In humans, it may be that the eating behaviour of some individuals resembles partial reinforcement schedules to a greater extent, possibly affecting dieting success by interacting with extinction and reacquisition. Furthermore, impulsivity has been associated with less successful dieting, and this association might be explained by impulsivity affecting the learning and extinction of appetitive responses. In the present two studies, the effects of different reinforcement schedules and impulsivity on the acquisition, extinction, and reacquisition of appetitive responses were investigated in a conditioning paradigm involving food rewards in healthy humans. Overall, the results indicate both partial reinforcement schedules and, possibly, impulsivity to be associated with worse extinction performance. A new model of dieting success is proposed: learning histories and, perhaps, certain personality traits (impulsivity) can interfere with the extinction and reacquisition of appetitive responses to food cues and they may be causally related to unsuccessful dieting. Copyright © 2014 Elsevier Ltd. All rights reserved.

Neural networks for continuous online learning and control.

PubMed

Choy, Min Chee; Srinivasan, Dipti; Cheu, Ruey Long

2006-11-01

This paper proposes a new hybrid neural network (NN) model that employs a multistage online learning process to solve the distributed control problem with an infinite horizon. Various techniques such as reinforcement learning and evolutionary algorithm are used to design the multistage online learning process. For this paper, the infinite horizon distributed control problem is implemented in the form of real-time distributed traffic signal control for intersections in a large-scale traffic network. The hybrid neural network model is used to design each of the local traffic signal controllers at the respective intersections. As the state of the traffic network changes due to random fluctuation of traffic volumes, the NN-based local controllers will need to adapt to the changing dynamics in order to provide effective traffic signal control and to prevent the traffic network from becoming overcongested. Such a problem is especially challenging if the local controllers are used for an infinite horizon problem where online learning has to take place continuously once the controllers are implemented into the traffic network. A comprehensive simulation model of a section of the Central Business District (CBD) of Singapore has been developed using PARAMICS microscopic simulation program. As the complexity of the simulation increases, results show that the hybrid NN model provides significant improvement in traffic conditions when evaluated against an existing traffic signal control algorithm as well as a new, continuously updated simultaneous perturbation stochastic approximation-based neural network (SPSA-NN). Using the hybrid NN model, the total mean delay of each vehicle has been reduced by 78% and the total mean stoppage time of each vehicle has been reduced by 84% compared to the existing traffic signal control algorithm. This shows the efficacy of the hybrid NN model in solving large-scale traffic signal control problem in a distributed manner. Also, it indicates the possibility of using the hybrid NN model for other applications that are similar in nature as the infinite horizon distributed control problem.

Regulating recognition decisions through incremental reinforcement learning.

PubMed

Han, Sanghoon; Dobbins, Ian G

2009-06-01

Does incremental reinforcement learning influence recognition memory judgments? We examined this question by subtly altering the relative validity or availability of feedback in order to differentially reinforce old or new recognition judgments. Experiment 1 probabilistically and incorrectly indicated that either misses or false alarms were correct in the context of feedback that was otherwise accurate. Experiment 2 selectively withheld feedback for either misses or false alarms in the context of feedback that was otherwise present. Both manipulations caused prominent shifts of recognition memory decision criteria that remained for considerable periods even after feedback had been altogether removed. Overall, these data demonstrate that incremental reinforcement-learning mechanisms influence the degree of caution subjects exercise when evaluating explicit memories.

Infant Contingency Learning in Different Cultural Contexts

ERIC Educational Resources Information Center

Graf, Frauke; Lamm, Bettina; Goertz, Claudia; Kolling, Thorsten; Freitag, Claudia; Spangler, Sibylle; Fassbender, Ina; Teubert, Manuel; Vierhaus, Marc; Keller, Heidi; Lohaus, Arnold; Schwarzer, Gudrun; Knopf, Monika

2012-01-01

Three-month-old Cameroonian Nso farmer and German middle-class infants were compared regarding learning and retention in a computerized mobile task. Infants achieving a preset learning criterion during reinforcement were tested for immediate and long-term retention measured in terms of an increased response rate after reinforcement and after a…

Adaptive Educational Software by Applying Reinforcement Learning

ERIC Educational Resources Information Center

Bennane, Abdellah

2013-01-01

The introduction of the intelligence in teaching software is the object of this paper. In software elaboration process, one uses some learning techniques in order to adapt the teaching software to characteristics of student. Generally, one uses the artificial intelligence techniques like reinforcement learning, Bayesian network in order to adapt…

A Robust Cooperated Control Method with Reinforcement Learning and Adaptive H∞ Control

NASA Astrophysics Data System (ADS)

Obayashi, Masanao; Uchiyama, Shogo; Kuremoto, Takashi; Kobayashi, Kunikazu

This study proposes a robust cooperated control method combining reinforcement learning with robust control to control the system. A remarkable characteristic of the reinforcement learning is that it doesn't require model formula, however, it doesn't guarantee the stability of the system. On the other hand, robust control system guarantees stability and robustness, however, it requires model formula. We employ both the actor-critic method which is a kind of reinforcement learning with minimal amount of computation to control continuous valued actions and the traditional robust control, that is, H∞ control. The proposed system was compared method with the conventional control method, that is, the actor-critic only used, through the computer simulation of controlling the angle and the position of a crane system, and the simulation result showed the effectiveness of the proposed method.

Mastering the game of Go with deep neural networks and tree search

NASA Astrophysics Data System (ADS)

Silver, David; Huang, Aja; Maddison, Chris J.; Guez, Arthur; Sifre, Laurent; van den Driessche, George; Schrittwieser, Julian; Antonoglou, Ioannis; Panneershelvam, Veda; Lanctot, Marc; Dieleman, Sander; Grewe, Dominik; Nham, John; Kalchbrenner, Nal; Sutskever, Ilya; Lillicrap, Timothy; Leach, Madeleine; Kavukcuoglu, Koray; Graepel, Thore; Hassabis, Demis

2016-01-01

The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses ‘value networks’ to evaluate board positions and ‘policy networks’ to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

Mastering the game of Go with deep neural networks and tree search.

PubMed

Silver, David; Huang, Aja; Maddison, Chris J; Guez, Arthur; Sifre, Laurent; van den Driessche, George; Schrittwieser, Julian; Antonoglou, Ioannis; Panneershelvam, Veda; Lanctot, Marc; Dieleman, Sander; Grewe, Dominik; Nham, John; Kalchbrenner, Nal; Sutskever, Ilya; Lillicrap, Timothy; Leach, Madeleine; Kavukcuoglu, Koray; Graepel, Thore; Hassabis, Demis

2016-01-28

The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses 'value networks' to evaluate board positions and 'policy networks' to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art Monte Carlo tree search programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines Monte Carlo simulation with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

Adaptive dynamic programming approach to experience-based systems identification and control.

PubMed

Lendaris, George G

2009-01-01

Humans have the ability to make use of experience while selecting their control actions for distinct and changing situations, and their process speeds up and have enhanced effectiveness as more experience is gained. In contrast, current technological implementations slow down as more knowledge is stored. A novel way of employing Approximate (or Adaptive) Dynamic Programming (ADP) is described that shifts the underlying Adaptive Critic type of Reinforcement Learning method "up a level", away from designing individual (optimal) controllers to that of developing on-line algorithms that efficiently and effectively select designs from a repository of existing controller solutions (perhaps previously developed via application of ADP methods). The resulting approach is called Higher-Level Learning Algorithm. The approach and its rationale are described and some examples of its application are given. The notions of context and context discernment are important to understanding the human abilities noted above. These are first defined, in a manner appropriate to controls and system-identification, and as a foundation relating to the application arena, a historical view of the various phases during development of the controls field is given, organized by how the notion 'context' was, or was not, involved in each phase.

Punishment insensitivity and impaired reinforcement learning in preschoolers.

PubMed

Briggs-Gowan, Margaret J; Nichols, Sara R; Voss, Joel; Zobel, Elvira; Carter, Alice S; McCarthy, Kimberly J; Pine, Daniel S; Blair, James; Wakschlag, Lauren S

2014-01-01

Youth and adults with psychopathic traits display disrupted reinforcement learning. Advances in measurement now enable examination of this association in preschoolers. The current study examines relations between reinforcement learning in preschoolers and parent ratings of reduced responsiveness to socialization, conceptualized as a developmental vulnerability to psychopathic traits. One hundred and fifty-seven preschoolers (mean age 4.7 ± 0.8 years) participated in a substudy that was embedded within a larger project. Children completed the 'Stars-in-Jars' task, which involved learning to select rewarded jars and avoid punished jars. Maternal report of responsiveness to socialization was assessed with the Punishment Insensitivity and Low Concern for Others scales of the Multidimensional Assessment of Preschool Disruptive Behavior (MAP-DB). Punishment Insensitivity, but not Low Concern for Others, was significantly associated with reinforcement learning in multivariate models that accounted for age and sex. Specifically, higher Punishment Insensitivity was associated with significantly lower overall performance and more errors on punished trials ('passive avoidance'). Impairments in reinforcement learning manifest in preschoolers who are high in maternal ratings of Punishment Insensitivity. If replicated, these findings may help to pinpoint the neurodevelopmental antecedents of psychopathic tendencies and suggest novel intervention targets beginning in early childhood. © 2013 The Authors. Journal of Child Psychology and Psychiatry © 2013 Association for Child and Adolescent Mental Health.

Rats bred for helplessness exhibit positive reinforcement learning deficits which are not alleviated by an antidepressant dose of the MAO-B inhibitor deprenyl.

PubMed

Schulz, Daniela; Henn, Fritz A; Petri, David; Huston, Joseph P

2016-08-04

Principles of negative reinforcement learning may play a critical role in the etiology and treatment of depression. We examined the integrity of positive reinforcement learning in congenitally helpless (cH) rats, an animal model of depression, using a random ratio schedule and a devaluation-extinction procedure. Furthermore, we tested whether an antidepressant dose of the monoamine oxidase (MAO)-B inhibitor deprenyl would reverse any deficits in positive reinforcement learning. We found that cH rats (n=9) were impaired in the acquisition of even simple operant contingencies, such as a fixed interval (FI) 20 schedule. cH rats exhibited no apparent deficits in appetite or reward sensitivity. They reacted to the devaluation of food in a manner consistent with a dose-response relationship. Reinforcer motivation as assessed by lever pressing across sessions with progressively decreasing reward probabilities was highest in congenitally non-helpless (cNH, n=10) rats as long as the reward probabilities remained relatively high. cNH compared to wild-type (n=10) rats were also more resistant to extinction across sessions. Compared to saline (n=5), deprenyl (n=5) reduced the duration of immobility of cH rats in the forced swimming test, indicative of antidepressant effects, but did not restore any deficits in the acquisition of a FI 20 schedule. We conclude that positive reinforcement learning was impaired in rats bred for helplessness, possibly due to motivational impairments but not deficits in reward sensitivity, and that deprenyl exerted antidepressant effects but did not reverse the deficits in positive reinforcement learning. Copyright © 2016 IBRO. Published by Elsevier Ltd. All rights reserved.

Antipsychotic dose modulates behavioral and neural responses to feedback during reinforcement learning in schizophrenia.

PubMed

Insel, Catherine; Reinen, Jenna; Weber, Jochen; Wager, Tor D; Jarskog, L Fredrik; Shohamy, Daphna; Smith, Edward E

2014-03-01

Schizophrenia is characterized by an abnormal dopamine system, and dopamine blockade is the primary mechanism of antipsychotic treatment. Consistent with the known role of dopamine in reward processing, prior research has demonstrated that patients with schizophrenia exhibit impairments in reward-based learning. However, it remains unknown how treatment with antipsychotic medication impacts the behavioral and neural signatures of reinforcement learning in schizophrenia. The goal of this study was to examine whether antipsychotic medication modulates behavioral and neural responses to prediction error coding during reinforcement learning. Patients with schizophrenia completed a reinforcement learning task while undergoing functional magnetic resonance imaging. The task consisted of two separate conditions in which participants accumulated monetary gain or avoided monetary loss. Behavioral results indicated that antipsychotic medication dose was associated with altered behavioral approaches to learning, such that patients taking higher doses of medication showed increased sensitivity to negative reinforcement. Higher doses of antipsychotic medication were also associated with higher learning rates (LRs), suggesting that medication enhanced sensitivity to trial-by-trial feedback. Neuroimaging data demonstrated that antipsychotic dose was related to differences in neural signatures of feedback prediction error during the loss condition. Specifically, patients taking higher doses of medication showed attenuated prediction error responses in the striatum and the medial prefrontal cortex. These findings indicate that antipsychotic medication treatment may influence motivational processes in patients with schizophrenia.

How we learn to make decisions: rapid propagation of reinforcement learning prediction errors in humans.

PubMed

Krigolson, Olav E; Hassall, Cameron D; Handy, Todd C

2014-03-01

Our ability to make decisions is predicated upon our knowledge of the outcomes of the actions available to us. Reinforcement learning theory posits that actions followed by a reward or punishment acquire value through the computation of prediction errors-discrepancies between the predicted and the actual reward. A multitude of neuroimaging studies have demonstrated that rewards and punishments evoke neural responses that appear to reflect reinforcement learning prediction errors [e.g., Krigolson, O. E., Pierce, L. J., Holroyd, C. B., & Tanaka, J. W. Learning to become an expert: Reinforcement learning and the acquisition of perceptual expertise. Journal of Cognitive Neuroscience, 21, 1833-1840, 2009; Bayer, H. M., & Glimcher, P. W. Midbrain dopamine neurons encode a quantitative reward prediction error signal. Neuron, 47, 129-141, 2005; O'Doherty, J. P. Reward representations and reward-related learning in the human brain: Insights from neuroimaging. Current Opinion in Neurobiology, 14, 769-776, 2004; Holroyd, C. B., & Coles, M. G. H. The neural basis of human error processing: Reinforcement learning, dopamine, and the error-related negativity. Psychological Review, 109, 679-709, 2002]. Here, we used the brain ERP technique to demonstrate that not only do rewards elicit a neural response akin to a prediction error but also that this signal rapidly diminished and propagated to the time of choice presentation with learning. Specifically, in a simple, learnable gambling task, we show that novel rewards elicited a feedback error-related negativity that rapidly decreased in amplitude with learning. Furthermore, we demonstrate the existence of a reward positivity at choice presentation, a previously unreported ERP component that has a similar timing and topography as the feedback error-related negativity that increased in amplitude with learning. The pattern of results we observed mirrored the output of a computational model that we implemented to compute reward prediction errors and the changes in amplitude of these prediction errors at the time of choice presentation and reward delivery. Our results provide further support that the computations that underlie human learning and decision-making follow reinforcement learning principles.

Microstimulation of the human substantia nigra alters reinforcement learning.

PubMed

Ramayya, Ashwin G; Misra, Amrit; Baltuch, Gordon H; Kahana, Michael J

2014-05-14

Animal studies have shown that substantia nigra (SN) dopaminergic (DA) neurons strengthen action-reward associations during reinforcement learning, but their role in human learning is not known. Here, we applied microstimulation in the SN of 11 patients undergoing deep brain stimulation surgery for the treatment of Parkinson's disease as they performed a two-alternative probability learning task in which rewards were contingent on stimuli, rather than actions. Subjects demonstrated decreased learning from reward trials that were accompanied by phasic SN microstimulation compared with reward trials without stimulation. Subjects who showed large decreases in learning also showed an increased bias toward repeating actions after stimulation trials; therefore, stimulation may have decreased learning by strengthening action-reward associations rather than stimulus-reward associations. Our findings build on previous studies implicating SN DA neurons in preferentially strengthening action-reward associations during reinforcement learning. Copyright © 2014 the authors 0270-6474/14/346887-09$15.00/0.

Human reinforcement learning subdivides structured action spaces by learning effector-specific values

PubMed Central

Gershman, Samuel J.; Pesaran, Bijan; Daw, Nathaniel D.

2009-01-01

Humans and animals are endowed with a large number of effectors. Although this enables great behavioral flexibility, it presents an equally formidable reinforcement learning problem of discovering which actions are most valuable, due to the high dimensionality of the action space. An unresolved question is how neural systems for reinforcement learning – such as prediction error signals for action valuation associated with dopamine and the striatum – can cope with this “curse of dimensionality.” We propose a reinforcement learning framework that allows for learned action valuations to be decomposed into effector-specific components when appropriate to a task, and test it by studying to what extent human behavior and BOLD activity can exploit such a decomposition in a multieffector choice task. Subjects made simultaneous decisions with their left and right hands and received separate reward feedback for each hand movement. We found that choice behavior was better described by a learning model that decomposed the values of bimanual movements into separate values for each effector, rather than a traditional model that treated the bimanual actions as unitary with a single value. A decomposition of value into effector-specific components was also observed in value-related BOLD signaling, in the form of lateralized biases in striatal correlates of prediction error and anticipatory value correlates in the intraparietal sulcus. These results suggest that the human brain can use decomposed value representations to “divide and conquer” reinforcement learning over high-dimensional action spaces. PMID:19864565

Human reinforcement learning subdivides structured action spaces by learning effector-specific values.

PubMed

Gershman, Samuel J; Pesaran, Bijan; Daw, Nathaniel D

2009-10-28

Humans and animals are endowed with a large number of effectors. Although this enables great behavioral flexibility, it presents an equally formidable reinforcement learning problem of discovering which actions are most valuable because of the high dimensionality of the action space. An unresolved question is how neural systems for reinforcement learning-such as prediction error signals for action valuation associated with dopamine and the striatum-can cope with this "curse of dimensionality." We propose a reinforcement learning framework that allows for learned action valuations to be decomposed into effector-specific components when appropriate to a task, and test it by studying to what extent human behavior and blood oxygen level-dependent (BOLD) activity can exploit such a decomposition in a multieffector choice task. Subjects made simultaneous decisions with their left and right hands and received separate reward feedback for each hand movement. We found that choice behavior was better described by a learning model that decomposed the values of bimanual movements into separate values for each effector, rather than a traditional model that treated the bimanual actions as unitary with a single value. A decomposition of value into effector-specific components was also observed in value-related BOLD signaling, in the form of lateralized biases in striatal correlates of prediction error and anticipatory value correlates in the intraparietal sulcus. These results suggest that the human brain can use decomposed value representations to "divide and conquer" reinforcement learning over high-dimensional action spaces.

Separation of time-based and trial-based accounts of the partial reinforcement extinction effect.

PubMed

Bouton, Mark E; Woods, Amanda M; Todd, Travis P

2014-01-01

Two appetitive conditioning experiments with rats examined time-based and trial-based accounts of the partial reinforcement extinction effect (PREE). In the PREE, the loss of responding that occurs in extinction is slower when the conditioned stimulus (CS) has been paired with a reinforcer on some of its presentations (partially reinforced) instead of every presentation (continuously reinforced). According to a time-based or "time-accumulation" view (e.g., Gallistel and Gibbon, 2000), the PREE occurs because the organism has learned in partial reinforcement to expect the reinforcer after a larger amount of time has accumulated in the CS over trials. In contrast, according to a trial-based view (e.g., Capaldi, 1967), the PREE occurs because the organism has learned in partial reinforcement to expect the reinforcer after a larger number of CS presentations. Experiment 1 used a procedure that equated partially and continuously reinforced groups on their expected times to reinforcement during conditioning. A PREE was still observed. Experiment 2 then used an extinction procedure that allowed time in the CS and the number of trials to accumulate differentially through extinction. The PREE was still evident when responding was examined as a function of expected time units to the reinforcer, but was eliminated when responding was examined as a function of expected trial units to the reinforcer. There was no evidence that the animal responded according to the ratio of time accumulated during the CS in extinction over the time in the CS expected before the reinforcer. The results thus favor a trial-based account over a time-based account of extinction and the PREE. This article is part of a Special Issue entitled: Associative and Temporal Learning. Copyright © 2013 Elsevier B.V. All rights reserved.

Electrophysiological correlates of reinforcement learning in young people with Tourette syndrome with and without co-occurring ADHD symptoms.

PubMed

Shephard, Elizabeth; Jackson, Georgina M; Groom, Madeleine J

2016-06-01

Altered reinforcement learning is implicated in the causes of Tourette syndrome (TS) and attention-deficit/hyperactivity disorder (ADHD). TS and ADHD frequently co-occur but how this affects reinforcement learning has not been investigated. We examined the ability of young people with TS (n=18), TS+ADHD (N=17), ADHD (n=13) and typically developing controls (n=20) to learn and reverse stimulus-response (S-R) associations based on positive and negative reinforcement feedback. We used a 2 (TS-yes, TS-no)×2 (ADHD-yes, ADHD-no) factorial design to assess the effects of TS, ADHD, and their interaction on behavioural (accuracy, RT) and event-related potential (stimulus-locked P3, feedback-locked P2, feedback-related negativity, FRN) indices of learning and reversing the S-R associations. TS was associated with intact learning and reversal performance and largely typical ERP amplitudes. ADHD was associated with lower accuracy during S-R learning and impaired reversal learning (significantly reduced accuracy and a trend for smaller P3 amplitude). The results indicate that co-occurring ADHD symptoms impair reversal learning in TS+ADHD. The implications of these findings for behavioural tic therapies are discussed. Copyright © 2016 ISDN. Published by Elsevier Ltd. All rights reserved.

Monocular depth perception using image processing and machine learning

NASA Astrophysics Data System (ADS)

Hombali, Apoorv; Gorde, Vaibhav; Deshpande, Abhishek

2011-10-01

This paper primarily exploits some of the more obscure, but inherent properties of camera and image to propose a simpler and more efficient way of perceiving depth. The proposed method involves the use of a single stationary camera at an unknown perspective and an unknown height to determine depth of an object on unknown terrain. In achieving so a direct correlation between a pixel in an image and the corresponding location in real space has to be formulated. First, a calibration step is undertaken whereby the equation of the plane visible in the field of view is calculated along with the relative distance between camera and plane by using a set of derived spatial geometrical relations coupled with a few intrinsic properties of the system. The depth of an unknown object is then perceived by first extracting the object under observation using a series of image processing steps followed by exploiting the aforementioned mapping of pixel and real space coordinate. The performance of the algorithm is greatly enhanced by the introduction of reinforced learning making the system independent of hardware and environment. Furthermore the depth calculation function is modified with a supervised learning algorithm giving consistent improvement in results. Thus, the system uses the experience in past and optimizes the current run successively. Using the above procedure a series of experiments and trials are carried out to prove the concept and its efficacy.

Locally optimal control under unknown dynamics with learnt cost function: application to industrial robot positioning

NASA Astrophysics Data System (ADS)

Guérin, Joris; Gibaru, Olivier; Thiery, Stéphane; Nyiri, Eric

2017-01-01

Recent methods of Reinforcement Learning have enabled to solve difficult, high dimensional, robotic tasks under unknown dynamics using iterative Linear Quadratic Gaussian control theory. These algorithms are based on building a local time-varying linear model of the dynamics from data gathered through interaction with the environment. In such tasks, the cost function is often expressed directly in terms of the state and control variables so that it can be locally quadratized to run the algorithm. If the cost is expressed in terms of other variables, a model is required to compute the cost function from the variables manipulated. We propose a method to learn the cost function directly from the data, in the same way as for the dynamics. This way, the cost function can be defined in terms of any measurable quantity and thus can be chosen more appropriately for the task to be carried out. With our method, any sensor information can be used to design the cost function. We demonstrate the efficiency of this method through simulating, with the V-REP software, the learning of a Cartesian positioning task on several industrial robots with different characteristics. The robots are controlled in joint space and no model is provided a priori. Our results are compared with another model free technique, consisting in writing the cost function as a state variable.

Embedded Incremental Feature Selection for Reinforcement Learning

DTIC Science & Technology

2012-05-01

Prior to this work, feature selection for reinforce- ment learning has focused on linear value function ap- proximation ( Kolter and Ng, 2009; Parr et al...InProceed- ings of the the 23rd International Conference on Ma- chine Learning, pages 449–456. Kolter , J. Z. and Ng, A. Y. (2009). Regularization and feature

Social Learning, Reinforcement and Crime: Evidence from Three European Cities

ERIC Educational Resources Information Center

Tittle, Charles R.; Antonaccio, Olena; Botchkovar, Ekaterina

2012-01-01

This study reports a cross-cultural test of Social Learning Theory using direct measures of social learning constructs and focusing on the causal structure implied by the theory. Overall, the results strongly confirm the main thrust of the theory. Prior criminal reinforcement and current crime-favorable definitions are highly related in all three…

Learning with incomplete information and the mathematical structure behind it.

PubMed

Kühn, Reimer; Stamatescu, Ion-Olimpiu

2007-07-01

We investigate the problem of learning with incomplete information as exemplified by learning with delayed reinforcement. We study a two phase learning scenario in which a phase of Hebbian associative learning based on momentary internal representations is supplemented by an 'unlearning' phase depending on a graded reinforcement signal. The reinforcement signal quantifies the success-rate globally for a number of learning steps in phase one, and 'unlearning' is indiscriminate with respect to associations learnt in that phase. Learning according to this model is studied via simulations and analytically within a student-teacher scenario for both single layer networks and, for a committee machine. Success and speed of learning depend on the ratio lambda of the learning rates used for the associative Hebbian learning phase and for the unlearning-correction in response to the reinforcement signal, respectively. Asymptotically perfect generalization is possible only, if this ratio exceeds a critical value lambda( c ), in which case the generalization error exhibits a power law decay with the number of examples seen by the student, with an exponent that depends in a non-universal manner on the parameter lambda. We find these features to be robust against a wide spectrum of modifications of microscopic modelling details. Two illustrative applications-one of a robot learning to navigate a field containing obstacles, and the problem of identifying a specific component in a collection of stimuli-are also provided.

Framing Reinforcement Learning from Human Reward: Reward Positivity, Temporal Discounting, Episodicity, and Performance

DTIC Science & Technology

2014-09-29

Framing Reinforcement Learning from Human Reward: Reward Positivity, Temporal Discounting, Episodicity , and Performance W. Bradley Knox...positive a trainer’s reward values are; temporal discounting, the extent to which future reward is discounted in value; episodicity , whether task...learning occurs in discrete learning episodes instead of one continuing session; and task performance, the agent’s performance on the task the trainer

Fuzzy Q-Learning for Generalization of Reinforcement Learning

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.

1996-01-01

Fuzzy Q-Learning, introduced earlier by the author, is an extension of Q-Learning into fuzzy environments. GARIC is a methodology for fuzzy reinforcement learning. In this paper, we introduce GARIC-Q, a new method for doing incremental Dynamic Programming using a society of intelligent agents which are controlled at the top level by Fuzzy Q-Learning and at the local level, each agent learns and operates based on GARIC. GARIC-Q improves the speed and applicability of Fuzzy Q-Learning through generalization of input space by using fuzzy rules and bridges the gap between Q-Learning and rule based intelligent systems.

Framework for robot skill learning using reinforcement learning

NASA Astrophysics Data System (ADS)

Wei, Yingzi; Zhao, Mingyang

2003-09-01

Robot acquiring skill is a process similar to human skill learning. Reinforcement learning (RL) is an on-line actor critic method for a robot to develop its skill. The reinforcement function has become the critical component for its effect of evaluating the action and guiding the learning process. We present an augmented reward function that provides a new way for RL controller to incorporate prior knowledge and experience into the RL controller. Also, the difference form of augmented reward function is considered carefully. The additional reward beyond conventional reward will provide more heuristic information for RL. In this paper, we present a strategy for the task of complex skill learning. Automatic robot shaping policy is to dissolve the complex skill into a hierarchical learning process. The new form of value function is introduced to attain smooth motion switching swiftly. We present a formal, but practical, framework for robot skill learning and also illustrate with an example the utility of method for learning skilled robot control on line.

Proactivity and Reinforcement: The Contingency of Social Behavior

ERIC Educational Resources Information Center

Williams, J. Sherwood; And Others

1976-01-01

This paper analyzes development of group structure in terms of the stimulus-sampling perspective. Learning is the continual sampling of possibilities, with those reinforced possibilities increasing in probability of occurance. This contingency learning approach is tested experimentally. (NG)

Mobile robots exploration through cnn-based reinforcement learning.

PubMed

Tai, Lei; Liu, Ming

2016-01-01

Exploration in an unknown environment is an elemental application for mobile robots. In this paper, we outlined a reinforcement learning method aiming for solving the exploration problem in a corridor environment. The learning model took the depth image from an RGB-D sensor as the only input. The feature representation of the depth image was extracted through a pre-trained convolutional-neural-networks model. Based on the recent success of deep Q-network on artificial intelligence, the robot controller achieved the exploration and obstacle avoidance abilities in several different simulated environments. It is the first time that the reinforcement learning is used to build an exploration strategy for mobile robots through raw sensor information.

Altered neural encoding of prediction errors in assault-related posttraumatic stress disorder.

PubMed

Ross, Marisa C; Lenow, Jennifer K; Kilts, Clinton D; Cisler, Josh M

2018-05-12

Posttraumatic stress disorder (PTSD) is widely associated with deficits in extinguishing learned fear responses, which relies on mechanisms of reinforcement learning (e.g., updating expectations based on prediction errors). However, the degree to which PTSD is associated with impairments in general reinforcement learning (i.e., outside of the context of fear stimuli) remains poorly understood. Here, we investigate brain and behavioral differences in general reinforcement learning between adult women with and without a current diagnosis of PTSD. 29 adult females (15 PTSD with exposure to assaultive violence, 14 controls) underwent a neutral reinforcement-learning task (i.e., two arm bandit task) during fMRI. We modeled participant behavior using different adaptations of the Rescorla-Wagner (RW) model and used Independent Component Analysis to identify timecourses for large-scale a priori brain networks. We found that an anticorrelated and risk sensitive RW model best fit participant behavior, with no differences in computational parameters between groups. Women in the PTSD group demonstrated significantly less neural encoding of prediction errors in both a ventral striatum/mPFC and anterior insula network compared to healthy controls. Weakened encoding of prediction errors in the ventral striatum/mPFC and anterior insula during a general reinforcement learning task, outside of the context of fear stimuli, suggests the possibility of a broader conceptualization of learning differences in PTSD than currently proposed in current neurocircuitry models of PTSD. Copyright © 2018 Elsevier Ltd. All rights reserved.

Automated Inattention and Fatigue Detection System in Distance Education for Elementary School Students

ERIC Educational Resources Information Center

Hwang, Kuo-An; Yang, Chia-Hao

2009-01-01

Most courses based on distance learning focus on the cognitive domain of learning. Because students are sometimes inattentive or tired, they may neglect the attention goal of learning. This study proposes an auto-detection and reinforcement mechanism for the distance-education system based on the reinforcement teaching strategy. If a student is…

When, What, and How Much to Reward in Reinforcement Learning-Based Models of Cognition

ERIC Educational Resources Information Center

Janssen, Christian P.; Gray, Wayne D.

2012-01-01

Reinforcement learning approaches to cognitive modeling represent task acquisition as learning to choose the sequence of steps that accomplishes the task while maximizing a reward. However, an apparently unrecognized problem for modelers is choosing when, what, and how much to reward; that is, when (the moment: end of trial, subtask, or some other…

Dissociating error-based and reinforcement-based loss functions during sensorimotor learning

PubMed Central

McGregor, Heather R.; Mohatarem, Ayman

2017-01-01

It has been proposed that the sensorimotor system uses a loss (cost) function to evaluate potential movements in the presence of random noise. Here we test this idea in the context of both error-based and reinforcement-based learning. In a reaching task, we laterally shifted a cursor relative to true hand position using a skewed probability distribution. This skewed probability distribution had its mean and mode separated, allowing us to dissociate the optimal predictions of an error-based loss function (corresponding to the mean of the lateral shifts) and a reinforcement-based loss function (corresponding to the mode). We then examined how the sensorimotor system uses error feedback and reinforcement feedback, in isolation and combination, when deciding where to aim the hand during a reach. We found that participants compensated differently to the same skewed lateral shift distribution depending on the form of feedback they received. When provided with error feedback, participants compensated based on the mean of the skewed noise. When provided with reinforcement feedback, participants compensated based on the mode. Participants receiving both error and reinforcement feedback continued to compensate based on the mean while repeatedly missing the target, despite receiving auditory, visual and monetary reinforcement feedback that rewarded hitting the target. Our work shows that reinforcement-based and error-based learning are separable and can occur independently. Further, when error and reinforcement feedback are in conflict, the sensorimotor system heavily weights error feedback over reinforcement feedback. PMID:28753634

Dissociating error-based and reinforcement-based loss functions during sensorimotor learning.

PubMed

Cashaback, Joshua G A; McGregor, Heather R; Mohatarem, Ayman; Gribble, Paul L

2017-07-01

It has been proposed that the sensorimotor system uses a loss (cost) function to evaluate potential movements in the presence of random noise. Here we test this idea in the context of both error-based and reinforcement-based learning. In a reaching task, we laterally shifted a cursor relative to true hand position using a skewed probability distribution. This skewed probability distribution had its mean and mode separated, allowing us to dissociate the optimal predictions of an error-based loss function (corresponding to the mean of the lateral shifts) and a reinforcement-based loss function (corresponding to the mode). We then examined how the sensorimotor system uses error feedback and reinforcement feedback, in isolation and combination, when deciding where to aim the hand during a reach. We found that participants compensated differently to the same skewed lateral shift distribution depending on the form of feedback they received. When provided with error feedback, participants compensated based on the mean of the skewed noise. When provided with reinforcement feedback, participants compensated based on the mode. Participants receiving both error and reinforcement feedback continued to compensate based on the mean while repeatedly missing the target, despite receiving auditory, visual and monetary reinforcement feedback that rewarded hitting the target. Our work shows that reinforcement-based and error-based learning are separable and can occur independently. Further, when error and reinforcement feedback are in conflict, the sensorimotor system heavily weights error feedback over reinforcement feedback.

Altered Risk-Based Decision Making following Adolescent Alcohol Use Results from an Imbalance in Reinforcement Learning in Rats

PubMed Central

Hart, Andrew S.; Collins, Anne L.; Bernstein, Ilene L.; Phillips, Paul E. M.

2012-01-01

Alcohol use during adolescence has profound and enduring consequences on decision-making under risk. However, the fundamental psychological processes underlying these changes are unknown. Here, we show that alcohol use produces over-fast learning for better-than-expected, but not worse-than-expected, outcomes without altering subjective reward valuation. We constructed a simple reinforcement learning model to simulate altered decision making using behavioral parameters extracted from rats with a history of adolescent alcohol use. Remarkably, the learning imbalance alone was sufficient to simulate the divergence in choice behavior observed between these groups of animals. These findings identify a selective alteration in reinforcement learning following adolescent alcohol use that can account for a robust change in risk-based decision making persisting into later life. PMID:22615989

Comparative learning theory and its application in the training of horses.

PubMed

Cooper, J J

1998-11-01

Training can best be explained as a process that occurs through stimulus-response-reinforcement chains, whereby animals are conditioned to associate cues in their environment, with specific behavioural responses and their rewarding consequences. Research into learning in horses has concentrated on their powers of discrimination and on primary positive reinforcement schedules, where the correct response is paired with a desirable consequence such as food. In contrast, a number of other learning processes that are used in training have been widely studied in other species, but have received little scientific investigation in the horse. These include: negative reinforcement, where performance of the correct response is followed by removal of, or decrease in, intensity of a unpleasant stimulus; punishment, where an incorrect response is paired with an undesirable consequence, but without consistent prior warning; secondary conditioning, where a natural primary reinforcer such as food is closely associated with an arbitrary secondary reinforcer such as vocal praise; and variable or partial conditioning, where once the correct response has been learnt, reinforcement is presented according to an intermittent schedule to increase resistance to extinction outside of training.

Reliability-based design optimization of reinforced concrete structures including soil-structure interaction using a discrete gravitational search algorithm and a proposed metamodel

NASA Astrophysics Data System (ADS)

Khatibinia, M.; Salajegheh, E.; Salajegheh, J.; Fadaee, M. J.

2013-10-01

A new discrete gravitational search algorithm (DGSA) and a metamodelling framework are introduced for reliability-based design optimization (RBDO) of reinforced concrete structures. The RBDO of structures with soil-structure interaction (SSI) effects is investigated in accordance with performance-based design. The proposed DGSA is based on the standard gravitational search algorithm (GSA) to optimize the structural cost under deterministic and probabilistic constraints. The Monte-Carlo simulation (MCS) method is considered as the most reliable method for estimating the probabilities of reliability. In order to reduce the computational time of MCS, the proposed metamodelling framework is employed to predict the responses of the SSI system in the RBDO procedure. The metamodel consists of a weighted least squares support vector machine (WLS-SVM) and a wavelet kernel function, which is called WWLS-SVM. Numerical results demonstrate the efficiency and computational advantages of DGSA and the proposed metamodel for RBDO of reinforced concrete structures.

The nature of sexual reinforcement.

PubMed Central

Crawford, L L; Holloway, K S; Domjan, M

1993-01-01

Sexual reinforcers are not part of a regulatory system involved in the maintenance of critical metabolic processes, they differ for males and females, they differ as a function of species and mating system, and they show ontogenetic and seasonal changes related to endocrine conditions. Exposure to a member of the opposite sex without copulation can be sufficient for sexual reinforcement. However, copulatory access is a stronger reinforcer, and copulatory opportunity can serve to enhance the reinforcing efficacy of stimulus features of a sexual partner. Conversely, under certain conditions, noncopulatory exposure serves to decrease reinforcer efficacy. Many common learning phenomena such as acquisition, extinction, discrimination learning, second-order conditioning, and latent inhibition have been demonstrated in sexual conditioning. These observations extend the generality of findings obtained with more conventional reinforcers, but the mechanisms of these effects and their gender and species specificity remain to be explored. PMID:8354970

Mesolimbic confidence signals guide perceptual learning in the absence of external feedback

PubMed Central

Guggenmos, Matthias; Wilbertz, Gregor; Hebart, Martin N; Sterzer, Philipp

2016-01-01

It is well established that learning can occur without external feedback, yet normative reinforcement learning theories have difficulties explaining such instances of learning. Here, we propose that human observers are capable of generating their own feedback signals by monitoring internal decision variables. We investigated this hypothesis in a visual perceptual learning task using fMRI and confidence reports as a measure for this monitoring process. Employing a novel computational model in which learning is guided by confidence-based reinforcement signals, we found that mesolimbic brain areas encoded both anticipation and prediction error of confidence—in remarkable similarity to previous findings for external reward-based feedback. We demonstrate that the model accounts for choice and confidence reports and show that the mesolimbic confidence prediction error modulation derived through the model predicts individual learning success. These results provide a mechanistic neurobiological explanation for learning without external feedback by augmenting reinforcement models with confidence-based feedback. DOI: http://dx.doi.org/10.7554/eLife.13388.001 PMID:27021283

Phasic dopamine as a prediction error of intrinsic and extrinsic reinforcements driving both action acquisition and reward maximization: a simulated robotic study.

PubMed

Mirolli, Marco; Santucci, Vieri G; Baldassarre, Gianluca

2013-03-01

An important issue of recent neuroscientific research is to understand the functional role of the phasic release of dopamine in the striatum, and in particular its relation to reinforcement learning. The literature is split between two alternative hypotheses: one considers phasic dopamine as a reward prediction error similar to the computational TD-error, whose function is to guide an animal to maximize future rewards; the other holds that phasic dopamine is a sensory prediction error signal that lets the animal discover and acquire novel actions. In this paper we propose an original hypothesis that integrates these two contrasting positions: according to our view phasic dopamine represents a TD-like reinforcement prediction error learning signal determined by both unexpected changes in the environment (temporary, intrinsic reinforcements) and biological rewards (permanent, extrinsic reinforcements). Accordingly, dopamine plays the functional role of driving both the discovery and acquisition of novel actions and the maximization of future rewards. To validate our hypothesis we perform a series of experiments with a simulated robotic system that has to learn different skills in order to get rewards. We compare different versions of the system in which we vary the composition of the learning signal. The results show that only the system reinforced by both extrinsic and intrinsic reinforcements is able to reach high performance in sufficiently complex conditions. Copyright © 2013 Elsevier Ltd. All rights reserved.

Within- and across-trial dynamics of human EEG reveal cooperative interplay between reinforcement learning and working memory.

PubMed

Collins, Anne G E; Frank, Michael J

2018-03-06

Learning from rewards and punishments is essential to survival and facilitates flexible human behavior. It is widely appreciated that multiple cognitive and reinforcement learning systems contribute to decision-making, but the nature of their interactions is elusive. Here, we leverage methods for extracting trial-by-trial indices of reinforcement learning (RL) and working memory (WM) in human electro-encephalography to reveal single-trial computations beyond that afforded by behavior alone. Neural dynamics confirmed that increases in neural expectation were predictive of reduced neural surprise in the following feedback period, supporting central tenets of RL models. Within- and cross-trial dynamics revealed a cooperative interplay between systems for learning, in which WM contributes expectations to guide RL, despite competition between systems during choice. Together, these results provide a deeper understanding of how multiple neural systems interact for learning and decision-making and facilitate analysis of their disruption in clinical populations.

Learning and tuning fuzzy logic controllers through reinforcements.

PubMed

Berenji, H R; Khedkar, P

1992-01-01

A method for learning and tuning a fuzzy logic controller based on reinforcements from a dynamic system is presented. It is shown that: the generalized approximate-reasoning-based intelligent control (GARIC) architecture learns and tunes a fuzzy logic controller even when only weak reinforcement, such as a binary failure signal, is available; introduces a new conjunction operator in computing the rule strengths of fuzzy control rules; introduces a new localized mean of maximum (LMOM) method in combining the conclusions of several firing control rules; and learns to produce real-valued control actions. Learning is achieved by integrating fuzzy inference into a feedforward network, which can then adaptively improve performance by using gradient descent methods. The GARIC architecture is applied to a cart-pole balancing system and demonstrates significant improvements in terms of the speed of learning and robustness to changes in the dynamic system's parameters over previous schemes for cart-pole balancing.

Impairments in action-outcome learning in schizophrenia.

PubMed

Morris, Richard W; Cyrzon, Chad; Green, Melissa J; Le Pelley, Mike E; Balleine, Bernard W

2018-03-03

Learning the causal relation between actions and their outcomes (AO learning) is critical for goal-directed behavior when actions are guided by desire for the outcome. This can be contrasted with habits that are acquired by reinforcement and primed by prevailing stimuli, in which causal learning plays no part. Recently, we demonstrated that goal-directed actions are impaired in schizophrenia; however, whether this deficit exists alongside impairments in habit or reinforcement learning is unknown. The present study distinguished deficits in causal learning from reinforcement learning in schizophrenia. We tested people with schizophrenia (SZ, n = 25) and healthy adults (HA, n = 25) in a vending machine task. Participants learned two action-outcome contingencies (e.g., push left to get a chocolate M&M, push right to get a cracker), and they also learned one contingency was degraded by delivery of noncontingent outcomes (e.g., free M&Ms), as well as changes in value by outcome devaluation. Both groups learned the best action to obtain rewards; however, SZ did not distinguish the more causal action when one AO contingency was degraded. Moreover, action selection in SZ was insensitive to changes in outcome value unless feedback was provided, and this was related to the deficit in AO learning. The failure to encode the causal relation between action and outcome in schizophrenia occurred without any apparent deficit in reinforcement learning. This implies that poor goal-directed behavior in schizophrenia cannot be explained by a more primary deficit in reward learning such as insensitivity to reward value or reward prediction errors.

The Effects of a Token Reinforcement System on the Reading and Arithmetic Skills Learnings of Migrant Primary School Pupils.

ERIC Educational Resources Information Center

Heitzman, Andrew J.

The New York State Center for Migrant Studies conducted this 1968 study which investigated effects of token reinforcers on reading and arithmetic skills learnings of migrant primary school students during a 6-week summer school session. Students (Negro and Caucasian) received plastic tokens to reward skills learning responses. Tokens were traded…

An Evaluation of Pedagogical Tutorial Tactics for a Natural Language Tutoring System: A Reinforcement Learning Approach

ERIC Educational Resources Information Center

Chi, Min; VanLehn, Kurt; Litman, Diane; Jordan, Pamela

2011-01-01

Pedagogical strategies are policies for a tutor to decide the next action when there are multiple actions available. When the content is controlled to be the same across experimental conditions, there has been little evidence that tutorial decisions have an impact on students' learning. In this paper, we applied Reinforcement Learning (RL) to…

The Identification and Establishment of Reinforcement for Collaboration in Elementary Students

ERIC Educational Resources Information Center

Darcy, Laura

2017-01-01

In Experiment 1, I conducted a functional analysis of student rate of learning with and without a peer-yoked contingency for 12 students in Kindergarten through 2nd grade in order to determine if they had conditioned reinforcement for collaboration. Using an ABAB reversal design, I compared rate of learning as measured by learn units to criterion…

Stress enhances model-free reinforcement learning only after negative outcome

PubMed Central

Lee, Daeyeol

2017-01-01

Previous studies found that stress shifts behavioral control by promoting habits while decreasing goal-directed behaviors during reward-based decision-making. It is, however, unclear how stress disrupts the relative contribution of the two systems controlling reward-seeking behavior, i.e. model-free (or habit) and model-based (or goal-directed). Here, we investigated whether stress biases the contribution of model-free and model-based reinforcement learning processes differently depending on the valence of outcome, and whether stress alters the learning rate, i.e., how quickly information from the new environment is incorporated into choices. Participants were randomly assigned to either a stress or a control condition, and performed a two-stage Markov decision-making task in which the reward probabilities underwent periodic reversals without notice. We found that stress increased the contribution of model-free reinforcement learning only after negative outcome. Furthermore, stress decreased the learning rate. The results suggest that stress diminishes one’s ability to make adaptive choices in multiple aspects of reinforcement learning. This finding has implications for understanding how stress facilitates maladaptive habits, such as addictive behavior, and other dysfunctional behaviors associated with stress in clinical and educational contexts. PMID:28723943

Stress enhances model-free reinforcement learning only after negative outcome.

PubMed

Park, Heyeon; Lee, Daeyeol; Chey, Jeanyung

2017-01-01

Previous studies found that stress shifts behavioral control by promoting habits while decreasing goal-directed behaviors during reward-based decision-making. It is, however, unclear how stress disrupts the relative contribution of the two systems controlling reward-seeking behavior, i.e. model-free (or habit) and model-based (or goal-directed). Here, we investigated whether stress biases the contribution of model-free and model-based reinforcement learning processes differently depending on the valence of outcome, and whether stress alters the learning rate, i.e., how quickly information from the new environment is incorporated into choices. Participants were randomly assigned to either a stress or a control condition, and performed a two-stage Markov decision-making task in which the reward probabilities underwent periodic reversals without notice. We found that stress increased the contribution of model-free reinforcement learning only after negative outcome. Furthermore, stress decreased the learning rate. The results suggest that stress diminishes one's ability to make adaptive choices in multiple aspects of reinforcement learning. This finding has implications for understanding how stress facilitates maladaptive habits, such as addictive behavior, and other dysfunctional behaviors associated with stress in clinical and educational contexts.

Implicit chaining in cotton-top tamarins (Saguinus oedipus) with elements equated for probability of reinforcement

PubMed Central

Dillon, Laura; Collins, Meaghan; Conway, Maura; Cunningham, Kate

2013-01-01

Three experiments examined the implicit learning of sequences under conditions in which the elements comprising a sequence were equated in terms of reinforcement probability. In Experiment 1 cotton-top tamarins (Saguinus oedipus) experienced a five-element sequence displayed serially on a touch screen in which reinforcement probability was equated across elements at .16 per element. Tamarins demonstrated learning of this sequence with higher latencies during a random test as compared to baseline sequence training. In Experiments 2 and 3, manipulations of the procedure used in the first experiment were undertaken to rule out a confound owing to the fact that the elements in Experiment 1 bore different temporal relations to the intertrial interval (ITI), an inhibitory period. The results of Experiments 2 and 3 indicated that the implicit learning observed in Experiment 1 was not due to temporal proximity between some elements and the inhibitory ITI. The results taken together support two conclusion: First that tamarins engaged in sequence learning whether or not there was contingent reinforcement for learning the sequence, and second that this learning was not due to subtle differences in associative strength between the elements of the sequence. PMID:23344718

Improving the Science Excursion: An Educational Technologist's View

ERIC Educational Resources Information Center

Balson, M.

1973-01-01

Analyzes the nature of the learning process and attempts to show how the three components of a reinforcement contingency, the stimulus, the response and the reinforcement can be utilized to increase the efficiency of a typical science learning experience, the excursion. (JR)

The algorithmic anatomy of model-based evaluation

PubMed Central

Daw, Nathaniel D.; Dayan, Peter

2014-01-01

Despite many debates in the first half of the twentieth century, it is now largely a truism that humans and other animals build models of their environments and use them for prediction and control. However, model-based (MB) reasoning presents severe computational challenges. Alternative, computationally simpler, model-free (MF) schemes have been suggested in the reinforcement learning literature, and have afforded influential accounts of behavioural and neural data. Here, we study the realization of MB calculations, and the ways that this might be woven together with MF values and evaluation methods. There are as yet mostly only hints in the literature as to the resulting tapestry, so we offer more preview than review. PMID:25267820

Gaze-contingent reinforcement learning reveals incentive value of social signals in young children and adults

PubMed Central

Smith, Tim J.; Senju, Atsushi

2017-01-01

While numerous studies have demonstrated that infants and adults preferentially orient to social stimuli, it remains unclear as to what drives such preferential orienting. It has been suggested that the learned association between social cues and subsequent reward delivery might shape such social orienting. Using a novel, spontaneous indication of reinforcement learning (with the use of a gaze contingent reward-learning task), we investigated whether children and adults' orienting towards social and non-social visual cues can be elicited by the association between participants' visual attention and a rewarding outcome. Critically, we assessed whether the engaging nature of the social cues influences the process of reinforcement learning. Both children and adults learned to orient more often to the visual cues associated with reward delivery, demonstrating that cue–reward association reinforced visual orienting. More importantly, when the reward-predictive cue was social and engaging, both children and adults learned the cue–reward association faster and more efficiently than when the reward-predictive cue was social but non-engaging. These new findings indicate that social engaging cues have a positive incentive value. This could possibly be because they usually coincide with positive outcomes in real life, which could partly drive the development of social orienting. PMID:28250186

Gaze-contingent reinforcement learning reveals incentive value of social signals in young children and adults.

PubMed

Vernetti, Angélina; Smith, Tim J; Senju, Atsushi

2017-03-15

While numerous studies have demonstrated that infants and adults preferentially orient to social stimuli, it remains unclear as to what drives such preferential orienting. It has been suggested that the learned association between social cues and subsequent reward delivery might shape such social orienting. Using a novel, spontaneous indication of reinforcement learning (with the use of a gaze contingent reward-learning task), we investigated whether children and adults' orienting towards social and non-social visual cues can be elicited by the association between participants' visual attention and a rewarding outcome. Critically, we assessed whether the engaging nature of the social cues influences the process of reinforcement learning. Both children and adults learned to orient more often to the visual cues associated with reward delivery, demonstrating that cue-reward association reinforced visual orienting. More importantly, when the reward-predictive cue was social and engaging, both children and adults learned the cue-reward association faster and more efficiently than when the reward-predictive cue was social but non-engaging. These new findings indicate that social engaging cues have a positive incentive value. This could possibly be because they usually coincide with positive outcomes in real life, which could partly drive the development of social orienting. © 2017 The Authors.

Learning and altering behaviours by reinforcement: neurocognitive differences between children and adults.

PubMed

Shephard, E; Jackson, G M; Groom, M J

2014-01-01

This study examined neurocognitive differences between children and adults in the ability to learn and adapt simple stimulus-response associations through feedback. Fourteen typically developing children (mean age=10.2) and 15 healthy adults (mean age=25.5) completed a simple task in which they learned to associate visually presented stimuli with manual responses based on performance feedback (acquisition phase), and then reversed and re-learned those associations following an unexpected change in reinforcement contingencies (reversal phase). Electrophysiological activity was recorded throughout task performance. We found no group differences in learning-related changes in performance (reaction time, accuracy) or in the amplitude of event-related potentials (ERPs) associated with stimulus processing (P3 ERP) or feedback processing (feedback-related negativity; FRN) during the acquisition phase. However, children's performance was significantly more disrupted by the reversal than adults and FRN amplitudes were significantly modulated by the reversal phase in children but not adults. These findings indicate that children have specific difficulties with reinforcement learning when acquired behaviours must be altered. This may be caused by the added demands on immature executive functioning, specifically response monitoring, created by the requirement to reverse the associations, or a developmental difference in the way in which children and adults approach reinforcement learning. Copyright © 2013 The Authors. Published by Elsevier Ltd.. All rights reserved.

Reinforcement Learning with Orthonormal Basis Adaptation Based on Activity-Oriented Index Allocation

NASA Astrophysics Data System (ADS)

Satoh, Hideki

An orthonormal basis adaptation method for function approximation was developed and applied to reinforcement learning with multi-dimensional continuous state space. First, a basis used for linear function approximation of a control function is set to an orthonormal basis. Next, basis elements with small activities are replaced with other candidate elements as learning progresses. As this replacement is repeated, the number of basis elements with large activities increases. Example chaos control problems for multiple logistic maps were solved, demonstrating that the method for adapting an orthonormal basis can modify a basis while holding the orthonormality in accordance with changes in the environment to improve the performance of reinforcement learning and to eliminate the adverse effects of redundant noisy states.

A neural model of hierarchical reinforcement learning.

PubMed

Rasmussen, Daniel; Voelker, Aaron; Eliasmith, Chris

2017-01-01

We develop a novel, biologically detailed neural model of reinforcement learning (RL) processes in the brain. This model incorporates a broad range of biological features that pose challenges to neural RL, such as temporally extended action sequences, continuous environments involving unknown time delays, and noisy/imprecise computations. Most significantly, we expand the model into the realm of hierarchical reinforcement learning (HRL), which divides the RL process into a hierarchy of actions at different levels of abstraction. Here we implement all the major components of HRL in a neural model that captures a variety of known anatomical and physiological properties of the brain. We demonstrate the performance of the model in a range of different environments, in order to emphasize the aim of understanding the brain's general reinforcement learning ability. These results show that the model compares well to previous modelling work and demonstrates improved performance as a result of its hierarchical ability. We also show that the model's behaviour is consistent with available data on human hierarchical RL, and generate several novel predictions.

Neural correlates of reinforcement learning and social preferences in competitive bidding.

PubMed

van den Bos, Wouter; Talwar, Arjun; McClure, Samuel M

2013-01-30

In competitive social environments, people often deviate from what rational choice theory prescribes, resulting in losses or suboptimal monetary gains. We investigate how competition affects learning and decision-making in a common value auction task. During the experiment, groups of five human participants were simultaneously scanned using MRI while playing the auction task. We first demonstrate that bidding is well characterized by reinforcement learning with biased reward representations dependent on social preferences. Indicative of reinforcement learning, we found that estimated trial-by-trial prediction errors correlated with activity in the striatum and ventromedial prefrontal cortex. Additionally, we found that individual differences in social preferences were related to activity in the temporal-parietal junction and anterior insula. Connectivity analyses suggest that monetary and social value signals are integrated in the ventromedial prefrontal cortex and striatum. Based on these results, we argue for a novel mechanistic account for the integration of reinforcement history and social preferences in competitive decision-making.

Attentional Selection Can Be Predicted by Reinforcement Learning of Task-relevant Stimulus Features Weighted by Value-independent Stickiness.

PubMed

Balcarras, Matthew; Ardid, Salva; Kaping, Daniel; Everling, Stefan; Womelsdorf, Thilo

2016-02-01

Attention includes processes that evaluate stimuli relevance, select the most relevant stimulus against less relevant stimuli, and bias choice behavior toward the selected information. It is not clear how these processes interact. Here, we captured these processes in a reinforcement learning framework applied to a feature-based attention task that required macaques to learn and update the value of stimulus features while ignoring nonrelevant sensory features, locations, and action plans. We found that value-based reinforcement learning mechanisms could account for feature-based attentional selection and choice behavior but required a value-independent stickiness selection process to explain selection errors while at asymptotic behavior. By comparing different reinforcement learning schemes, we found that trial-by-trial selections were best predicted by a model that only represents expected values for the task-relevant feature dimension, with nonrelevant stimulus features and action plans having only a marginal influence on covert selections. These findings show that attentional control subprocesses can be described by (1) the reinforcement learning of feature values within a restricted feature space that excludes irrelevant feature dimensions, (2) a stochastic selection process on feature-specific value representations, and (3) value-independent stickiness toward previous feature selections akin to perseveration in the motor domain. We speculate that these three mechanisms are implemented by distinct but interacting brain circuits and that the proposed formal account of feature-based stimulus selection will be important to understand how attentional subprocesses are implemented in primate brain networks.

Extinction of Pavlovian conditioning: The influence of trial number and reinforcement history.

PubMed

Chan, C K J; Harris, Justin A

2017-08-01

Pavlovian conditioning is sensitive to the temporal relationship between the conditioned stimulus (CS) and the unconditioned stimulus (US). This has motivated models that describe learning as a process that continuously updates associative strength during the trial or specifically encodes the CS-US interval. These models predict that extinction of responding is also continuous, such that response loss is proportional to the cumulative duration of exposure to the CS without the US. We review evidence showing that this prediction is incorrect, and that extinction is trial-based rather than time-based. We also present two experiments that test the importance of trials versus time on the Partial Reinforcement Extinction Effect (PREE), in which responding extinguishes more slowly for a CS that was inconsistently reinforced with the US than for a consistently reinforced one. We show that increasing the number of extinction trials of the partially reinforced CS, relative to the consistently reinforced CS, overcomes the PREE. However, increasing the duration of extinction trials by the same amount does not overcome the PREE. We conclude that animals learn about the likelihood of the US per trial during conditioning, and learn trial-by-trial about the absence of the US during extinction. Moreover, what they learn about the likelihood of the US during conditioning affects how sensitive they are to the absence of the US during extinction. Copyright © 2017 Elsevier B.V. All rights reserved.

Can Service Learning Reinforce Social and Cultural Bias? Exploring a Popular Model of Family Involvement for Early Childhood Teacher Candidates

ERIC Educational Resources Information Center

Dunn-Kenney, Maylan

2010-01-01

Service learning is often used in teacher education as a way to challenge social bias and provide teacher candidates with skills needed to work in partnership with diverse families. Although some literature suggests that service learning could reinforce cultural bias, there is little documentation. In a study of 21 early childhood teacher…

Deep Gate Recurrent Neural Network

DTIC Science & Technology

2016-11-22

Schmidhuber. A system for robotic heart surgery that learns to tie knots using recurrent neural networks. In IEEE International Conference on...tasks, such as Machine Translation (Bahdanau et al. (2015)) or Robot Reinforcement Learning (Bakker (2001)). The main idea behind these networks is to...and J. Peters. Reinforcement learning in robotics : A survey. The International Journal of Robotics Research, 32:1238–1274, 2013. ISSN 0278-3649. doi

An Actor-Critic based controller for glucose regulation in type 1 diabetes.

PubMed

Daskalaki, Elena; Diem, Peter; Mougiakakou, Stavroula G

2013-02-01

A novel adaptive approach for glucose control in individuals with type 1 diabetes under sensor-augmented pump therapy is proposed. The controller, is based on Actor-Critic (AC) learning and is inspired by the principles of reinforcement learning and optimal control theory. The main characteristics of the proposed controller are (i) simultaneous adjustment of both the insulin basal rate and the bolus dose, (ii) initialization based on clinical procedures, and (iii) real-time personalization. The effectiveness of the proposed algorithm in terms of glycemic control has been investigated in silico in adults, adolescents and children under open-loop and closed-loop approaches, using announced meals with uncertainties in the order of ±25% in the estimation of carbohydrates. The results show that glucose regulation is efficient in all three groups of patients, even with uncertainties in the level of carbohydrates in the meal. The percentages in the A+B zones of the Control Variability Grid Analysis (CVGA) were 100% for adults, and 93% for both adolescents and children. The AC based controller seems to be a promising approach for the automatic adjustment of insulin infusion in order to improve glycemic control. After optimization of the algorithm, the controller will be tested in a clinical trial. Copyright © 2012 Elsevier Ireland Ltd. All rights reserved.

Conceptualizing withdrawal-induced escalation of alcohol self-administration as a learned, plasticity-dependent process

PubMed Central

Walker, Brendan M.

2013-01-01

This article represents one of five contributions focusing on the topic “Plasticity and neuroadaptive responses within the extended amygdala in response to chronic or excessive alcohol exposure” that were developed by awardees participating in the Young Investigator Award Symposium at the “Alcoholism and Stress: A Framework for Future Treatment Strategies” conference in Volterra, Italy on May 3–6, 2011 that was organized/chaired by Drs. Antonio Noronha and Fulton Crews and sponsored by the National Institute on Alcohol Abuse and Alcoholism. This review discusses the dependence-induced neuroadaptations in affective systems that provide a basis for negative reinforcement learning and presents evidence demonstrating that escalated alcohol consumption during withdrawal is a learned, plasticity-dependent process. The review concludes by identifying changes within extended amygdala dynorphin/kappa-opioid receptor systems that could serve as the foundation for the occurrence of negative reinforcement processes. While some evidence contained herein may be specific to alcohol dependence-related learning and plasticity, much of the information will be of relevance to any addictive disorder involving negative reinforcement mechanisms. Collectively, the information presented within this review provides a framework to assess the negative reinforcing effects of alcohol in a manner that distinguishes neuroadaptations produced by chronic alcohol exposure from the actual plasticity that is associated with negative reinforcement learning in dependent organisms. PMID:22459874

Reinforcement Learning Strategies for Clinical Trials in Non-small Cell Lung Cancer

PubMed Central

Zhao, Yufan; Zeng, Donglin; Socinski, Mark A.; Kosorok, Michael R.

2010-01-01

Summary Typical regimens for advanced metastatic stage IIIB/IV non-small cell lung cancer (NSCLC) consist of multiple lines of treatment. We present an adaptive reinforcement learning approach to discover optimal individualized treatment regimens from a specially designed clinical trial (a “clinical reinforcement trial”) of an experimental treatment for patients with advanced NSCLC who have not been treated previously with systemic therapy. In addition to the complexity of the problem of selecting optimal compounds for first and second-line treatments based on prognostic factors, another primary goal is to determine the optimal time to initiate second-line therapy, either immediately or delayed after induction therapy, yielding the longest overall survival time. A reinforcement learning method called Q-learning is utilized which involves learning an optimal regimen from patient data generated from the clinical reinforcement trial. Approximating the Q-function with time-indexed parameters can be achieved by using a modification of support vector regression which can utilize censored data. Within this framework, a simulation study shows that the procedure can extract optimal regimens for two lines of treatment directly from clinical data without prior knowledge of the treatment effect mechanism. In addition, we demonstrate that the design reliably selects the best initial time for second-line therapy while taking into account the heterogeneity of NSCLC across patients. PMID:21385164

Depression, Activity, and Evaluation of Reinforcement

ERIC Educational Resources Information Center

Hammen, Constance L.; Glass, David R., Jr.

1975-01-01

This research attempted to find the causal relation between mood and level of reinforcement. An effort was made to learn what mood change might occur if depressed subjects increased their levels of participation in reinforcing activities. (Author/RK)

What Can Reinforcement Learning Teach Us About Non-Equilibrium Quantum Dynamics

NASA Astrophysics Data System (ADS)

Bukov, Marin; Day, Alexandre; Sels, Dries; Weinberg, Phillip; Polkovnikov, Anatoli; Mehta, Pankaj

Equilibrium thermodynamics and statistical physics are the building blocks of modern science and technology. Yet, our understanding of thermodynamic processes away from equilibrium is largely missing. In this talk, I will reveal the potential of what artificial intelligence can teach us about the complex behaviour of non-equilibrium systems. Specifically, I will discuss the problem of finding optimal drive protocols to prepare a desired target state in quantum mechanical systems by applying ideas from Reinforcement Learning [one can think of Reinforcement Learning as the study of how an agent (e.g. a robot) can learn and perfect a given policy through interactions with an environment.]. The driving protocols learnt by our agent suggest that the non-equilibrium world features possibilities easily defying intuition based on equilibrium physics.

Kinesthetic Reinforcement-Is It a Boon to Learning?

ERIC Educational Resources Information Center

Bohrer, Roxilu K.

1970-01-01

Language instruction, particularly in the elementary school, should be reinforced through the use of visual aids and through associated physical activity. Kinesthetic experiences provide an opportunity to make use of non-verbal cues to meaning, enliven classroom activities, and maximize learning for pupils. The author discusses the educational…

Reinforcing Basic Skills Through Social Studies. Grades 4-7.

ERIC Educational Resources Information Center

Lewis, Teresa Marie

Arranged into seven parts, this document provides a variety of games and activities, bulletin board ideas, overhead transparencies, student handouts, and learning station ideas to help reinforce basic social studies skills in the intermediate grades. In part 1, students learn about timelines, first constructing their own life timeline, then a…

Effects of Reinforcement on Peer Imitation in a Small Group Play Context

ERIC Educational Resources Information Center

Barton, Erin E.; Ledford, Jennifer R.

2018-01-01

Children with disabilities often have deficits in imitation skills, particularly in imitating peers. Imitation is considered a behavioral cusp--which, once learned, allows a child to access additional and previously unavailable learning opportunities. In the current study, researchers examined the efficacy of contingent reinforcement delivered…

Closed-loop and robust control of quantum systems.

PubMed

Chen, Chunlin; Wang, Lin-Cheng; Wang, Yuanlong

2013-01-01

For most practical quantum control systems, it is important and difficult to attain robustness and reliability due to unavoidable uncertainties in the system dynamics or models. Three kinds of typical approaches (e.g., closed-loop learning control, feedback control, and robust control) have been proved to be effective to solve these problems. This work presents a self-contained survey on the closed-loop and robust control of quantum systems, as well as a brief introduction to a selection of basic theories and methods in this research area, to provide interested readers with a general idea for further studies. In the area of closed-loop learning control of quantum systems, we survey and introduce such learning control methods as gradient-based methods, genetic algorithms (GA), and reinforcement learning (RL) methods from a unified point of view of exploring the quantum control landscapes. For the feedback control approach, the paper surveys three control strategies including Lyapunov control, measurement-based control, and coherent-feedback control. Then such topics in the field of quantum robust control as H(∞) control, sliding mode control, quantum risk-sensitive control, and quantum ensemble control are reviewed. The paper concludes with a perspective of future research directions that are likely to attract more attention.

Cost optimization of reinforced concrete cantilever retaining walls under seismic loading using a biogeography-based optimization algorithm with Levy flights

NASA Astrophysics Data System (ADS)

Aydogdu, Ibrahim

2017-03-01

In this article, a new version of a biogeography-based optimization algorithm with Levy flight distribution (LFBBO) is introduced and used for the optimum design of reinforced concrete cantilever retaining walls under seismic loading. The cost of the wall is taken as an objective function, which is minimized under the constraints implemented by the American Concrete Institute (ACI 318-05) design code and geometric limitations. The influence of peak ground acceleration (PGA) on optimal cost is also investigated. The solution of the problem is attained by the LFBBO algorithm, which is developed by adding Levy flight distribution to the mutation part of the biogeography-based optimization (BBO) algorithm. Five design examples, of which two are used in literature studies, are optimized in the study. The results are compared to test the performance of the LFBBO and BBO algorithms, to determine the influence of the seismic load and PGA on the optimal cost of the wall.

Harnessing wake vortices for efficient collective swimming via deep reinfrcement learning

NASA Astrophysics Data System (ADS)

Verma, Siddartha; Novati, Guido; Koumoutsakos, Petros; ChairComputing Science Team

2017-11-01

Collective motion may bestow evolutionary advantages to a number of animal species. Soaring flocks of birds, teeming swarms of insects, and swirling masses of schooling fish, all to some extent enjoy anti-predator benefits, increased foraging success, and enhanced problem-solving abilities. Coordinated activity may also provide energetic benefits, as in the case of large groups of fish where swimmers exploit unsteady flow-patterns generated in the wake. Both experimental and computational investigations of such scenarios are hampered by difficulties associated with studying multiple swimmers. Consequentially, the precise energy-saving mechanisms at play remain largely unknown. We combine high-fidelity numerical simulations of multiple, self propelled swimmers with novel deep reinforcement learning algorithms to discover optimal ways for swimmers to interact with unsteady wakes, in a fully unsupervised manner. We identify optimal flow-interaction strategies devised by the resulting autonomous swimmers, and use it to formulate an effective control-logic. We demonstrate, via 3D simulations of controlled groups that swimmers exploiting the learned strategy exhibit a significant reduction in energy-expenditure. ERC Advanced Investigator Award 341117.

Neurofeedback in Learning Disabled Children: Visual versus Auditory Reinforcement.

PubMed

Fernández, Thalía; Bosch-Bayard, Jorge; Harmony, Thalía; Caballero, María I; Díaz-Comas, Lourdes; Galán, Lídice; Ricardo-Garcell, Josefina; Aubert, Eduardo; Otero-Ojeda, Gloria

2016-03-01

Children with learning disabilities (LD) frequently have an EEG characterized by an excess of theta and a deficit of alpha activities. NFB using an auditory stimulus as reinforcer has proven to be a useful tool to treat LD children by positively reinforcing decreases of the theta/alpha ratio. The aim of the present study was to optimize the NFB procedure by comparing the efficacy of visual (with eyes open) versus auditory (with eyes closed) reinforcers. Twenty LD children with an abnormally high theta/alpha ratio were randomly assigned to the Auditory or the Visual group, where a 500 Hz tone or a visual stimulus (a white square), respectively, was used as a positive reinforcer when the value of the theta/alpha ratio was reduced. Both groups had signs consistent with EEG maturation, but only the Auditory Group showed behavioral/cognitive improvements. In conclusion, the auditory reinforcer was more efficacious in reducing the theta/alpha ratio, and it improved the cognitive abilities more than the visual reinforcer.

Methodology of shell structure reinforcement layout optimization

NASA Astrophysics Data System (ADS)

Szafrański, Tomasz; Małachowski, Jerzy; Damaziak, Krzysztof

2018-01-01

This paper presents an optimization process of a reinforced shell diffuser intended for a small wind turbine (rated power of 3 kW). The diffuser structure consists of multiple reinforcement and metal skin. This kind of structure is suitable for optimization in terms of selection of reinforcement density, stringers cross sections, sheet thickness, etc. The optimisation approach assumes the reduction of the amount of work to be done between the optimization process and the final product design. The proposed optimization methodology is based on application of a genetic algorithm to generate the optimal reinforcement layout. The obtained results are the basis for modifying the existing Small Wind Turbine (SWT) design.

The probability of reinforcement per trial affects posttrial responding and subsequent extinction but not within-trial responding.

PubMed

Harris, Justin A; Kwok, Dorothy W S

2018-01-01

During magazine approach conditioning, rats do not discriminate between a conditional stimulus (CS) that is consistently reinforced with food and a CS that is occasionally (partially) reinforced, as long as the CSs have the same overall reinforcement rate per second. This implies that rats are indifferent to the probability of reinforcement per trial. However, in the same rats, the per-trial reinforcement rate will affect subsequent extinction-responding extinguishes more rapidly for a CS that was consistently reinforced than for a partially reinforced CS. Here, we trained rats with consistently and partially reinforced CSs that were matched for overall reinforcement rate per second. We measured conditioned responding both during and immediately after the CSs. Differences in the per-trial probability of reinforcement did not affect the acquisition of responding during the CS but did affect subsequent extinction of that responding, and also affected the post-CS response rates during conditioning. Indeed, CSs with the same probability of reinforcement per trial evoked the same amount of post-CS responding even when they differed in overall reinforcement rate and thus evoked different amounts of responding during the CS. We conclude that reinforcement rate per second controls rats' acquisition of responding during the CS, but at the same time, rats also learn specifically about the probability of reinforcement per trial. The latter learning affects the rats' expectation of reinforcement as an outcome of the trial, which influences their ability to detect retrospectively that an opportunity for reinforcement was missed, and, in turn, drives extinction. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Establishment and Maintenance of Socially Learned Conditioned Reinforcement in Young Children: Elimination of the Role of Adults and View of Peers' Faces

ERIC Educational Resources Information Center

Zrinzo, Michelle; Greer, R. Douglas

2013-01-01

Prior research has demonstrated the establishment of reinforcers for learning and maintenance with young children as a function of social learning where a peer and an adult experimenter were present. The presence of an adult experimenter was eliminated in the present study to test if the effect produced in the prior studies would occur with only…

Structure identification in fuzzy inference using reinforcement learning

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.; Khedkar, Pratap

1993-01-01

In our previous work on the GARIC architecture, we have shown that the system can start with surface structure of the knowledge base (i.e., the linguistic expression of the rules) and learn the deep structure (i.e., the fuzzy membership functions of the labels used in the rules) by using reinforcement learning. Assuming the surface structure, GARIC refines the fuzzy membership functions used in the consequents of the rules using a gradient descent procedure. This hybrid fuzzy logic and reinforcement learning approach can learn to balance a cart-pole system and to backup a truck to its docking location after a few trials. In this paper, we discuss how to do structure identification using reinforcement learning in fuzzy inference systems. This involves identifying both surface as well as deep structure of the knowledge base. The term set of fuzzy linguistic labels used in describing the values of each control variable must be derived. In this process, splitting a label refers to creating new labels which are more granular than the original label and merging two labels creates a more general label. Splitting and merging of labels directly transform the structure of the action selection network used in GARIC by increasing or decreasing the number of hidden layer nodes.

Spared internal but impaired external reward prediction error signals in major depressive disorder during reinforcement learning.

PubMed

Bakic, Jasmina; Pourtois, Gilles; Jepma, Marieke; Duprat, Romain; De Raedt, Rudi; Baeken, Chris

2017-01-01

Major depressive disorder (MDD) creates debilitating effects on a wide range of cognitive functions, including reinforcement learning (RL). In this study, we sought to assess whether reward processing as such, or alternatively the complex interplay between motivation and reward might potentially account for the abnormal reward-based learning in MDD. A total of 35 treatment resistant MDD patients and 44 age matched healthy controls (HCs) performed a standard probabilistic learning task. RL was titrated using behavioral, computational modeling and event-related brain potentials (ERPs) data. MDD patients showed comparable learning rate compared to HCs. However, they showed decreased lose-shift responses as well as blunted subjective evaluations of the reinforcers used during the task, relative to HCs. Moreover, MDD patients showed normal internal (at the level of error-related negativity, ERN) but abnormal external (at the level of feedback-related negativity, FRN) reward prediction error (RPE) signals during RL, selectively when additional efforts had to be made to establish learning. Collectively, these results lend support to the assumption that MDD does not impair reward processing per se during RL. Instead, it seems to alter the processing of the emotional value of (external) reinforcers during RL, when additional intrinsic motivational processes have to be engaged. © 2016 Wiley Periodicals, Inc.

Optimization of a Lunar Pallet Lander Reinforcement Structure Using a Genetic Algorithm

NASA Technical Reports Server (NTRS)

Burt, Adam

2014-01-01

In this paper, a unique system level spacecraft design optimization will be presented. A Genetic Algorithm is used to design the global pattern of the reinforcing structure, while a gradient routine is used to adequately stiffen the sub-structure. The system level structural design includes determining the optimal physical location (and number) of reinforcing beams of a lunar pallet lander deck structure. Design of the substructure includes determining placement of secondary stiffeners and the number of rivets required for assembly.. In this optimization, several considerations are taken into account. The primary objective was to raise the primary natural frequencies of the structure such that the Pallet Lander primary structure does not significantly couple with the launch vehicle. A secondary objective is to determine how to properly stiffen the reinforcing beams so that the beam web resists the shear buckling load imparted by the spacecraft components mounted to the pallet lander deck during launch and landing. A third objective is that the calculated stress does not exceed the allowable strength of the material. These design requirements must be met while, minimizing the overall mass of the spacecraft. The final paper will discuss how the optimization was implemented as well as the results. While driven by optimization algorithms, the primary purpose of this effort was to demonstrate the capability of genetic algorithms to enable design automation in the preliminary design cycle. By developing a routine that can automatically generate designs through the use of Finite Element Analysis, considerable design efficiencies, both in time and overall product, can be obtained over more traditional brute force design methods.

Relationship between Reinforcement and Eye Movements during Ocular Motor Training with Learning Disabled Children.

ERIC Educational Resources Information Center

Punnett, Audrey F.; Steinhauer, Gene D.

1984-01-01

Four reading disabled children were given eight sessions of ocular motor training with reinforcement and eight sessions without reinforcement. Two reading disabled control Ss were treated similarly but received no ocular motor training. Results demonstrated that reinforcement can improve ocular motor skills, which in turn elevates reading…

Learning the specific quality of taste reinforcement in larval Drosophila.

PubMed

Schleyer, Michael; Miura, Daisuke; Tanimura, Teiichi; Gerber, Bertram

2015-01-27

The only property of reinforcement insects are commonly thought to learn about is its value. We show that larval Drosophila not only remember the value of reinforcement (How much?), but also its quality (What?). This is demonstrated both within the appetitive domain by using sugar vs amino acid as different reward qualities, and within the aversive domain by using bitter vs high-concentration salt as different qualities of punishment. From the available literature, such nuanced memories for the quality of reinforcement are unexpected and pose a challenge to present models of how insect memory is organized. Given that animals as simple as larval Drosophila, endowed with but 10,000 neurons, operate with both reinforcement value and quality, we suggest that both are fundamental aspects of mnemonic processing-in any brain.

The evolution of continuous learning of the structure of the environment

PubMed Central

Kolodny, Oren; Edelman, Shimon; Lotem, Arnon

2014-01-01

Continuous, ‘always on’, learning of structure from a stream of data is studied mainly in the fields of machine learning or language acquisition, but its evolutionary roots may go back to the first organisms that were internally motivated to learn and represent their environment. Here, we study under what conditions such continuous learning (CL) may be more adaptive than simple reinforcement learning and examine how it could have evolved from the same basic associative elements. We use agent-based computer simulations to compare three learning strategies: simple reinforcement learning; reinforcement learning with chaining (RL-chain) and CL that applies the same associative mechanisms used by the other strategies, but also seeks statistical regularities in the relations among all items in the environment, regardless of the initial association with food. We show that a sufficiently structured environment favours the evolution of both RL-chain and CL and that CL outperforms the other strategies when food is relatively rare and the time for learning is limited. This advantage of internally motivated CL stems from its ability to capture statistical patterns in the environment even before they are associated with food, at which point they immediately become useful for planning. PMID:24402920

The partial-reinforcement extinction effect and the contingent-sampling hypothesis.

PubMed

Hochman, Guy; Erev, Ido

2013-12-01

The partial-reinforcement extinction effect (PREE) implies that learning under partial reinforcements is more robust than learning under full reinforcements. While the advantages of partial reinforcements have been well-documented in laboratory studies, field research has failed to support this prediction. In the present study, we aimed to clarify this pattern. Experiment 1 showed that partial reinforcements increase the tendency to select the promoted option during extinction; however, this effect is much smaller than the negative effect of partial reinforcements on the tendency to select the promoted option during the training phase. Experiment 2 demonstrated that the overall effect of partial reinforcements varies inversely with the attractiveness of the alternative to the promoted behavior: The overall effect is negative when the alternative is relatively attractive, and positive when the alternative is relatively unattractive. These results can be captured with a contingent-sampling model assuming that people select options that provided the best payoff in similar past experiences. The best fit was obtained under the assumption that similarity is defined by the sequence of the last four outcomes.

The effects of aging on the interaction between reinforcement learning and attention.

PubMed

Radulescu, Angela; Daniel, Reka; Niv, Yael

2016-11-01

Reinforcement learning (RL) in complex environments relies on selective attention to uncover those aspects of the environment that are most predictive of reward. Whereas previous work has focused on age-related changes in RL, it is not known whether older adults learn differently from younger adults when selective attention is required. In 2 experiments, we examined how aging affects the interaction between RL and selective attention. Younger and older adults performed a learning task in which only 1 stimulus dimension was relevant to predicting reward, and within it, 1 "target" feature was the most rewarding. Participants had to discover this target feature through trial and error. In Experiment 1, stimuli varied on 1 or 3 dimensions and participants received hints that revealed the target feature, the relevant dimension, or gave no information. Group-related differences in accuracy and RTs differed systematically as a function of the number of dimensions and the type of hint available. In Experiment 2 we used trial-by-trial computational modeling of the learning process to test for age-related differences in learning strategies. Behavior of both young and older adults was explained well by a reinforcement-learning model that uses selective attention to constrain learning. However, the model suggested that older adults restricted their learning to fewer features, employing more focused attention than younger adults. Furthermore, this difference in strategy predicted age-related deficits in accuracy. We discuss these results suggesting that a narrower filter of attention may reflect an adaptation to the reduced capabilities of the reinforcement learning system. (PsycINFO Database Record (c) 2016 APA, all rights reserved).

Tiger salamanders' (Ambystoma tigrinum) response learning and usage of visual cues.

PubMed

Kundey, Shannon M A; Millar, Roberto; McPherson, Justin; Gonzalez, Maya; Fitz, Aleyna; Allen, Chadbourne

2016-05-01

We explored tiger salamanders' (Ambystoma tigrinum) learning to execute a response within a maze as proximal visual cue conditions varied. In Experiment 1, salamanders learned to turn consistently in a T-maze for reinforcement before the maze was rotated. All learned the initial task and executed the trained turn during test, suggesting that they learned to demonstrate the reinforced response during training and continued to perform it during test. In a second experiment utilizing a similar procedure, two visual cues were placed consistently at the maze junction. Salamanders were reinforced for turning towards one cue. Cue placement was reversed during test. All learned the initial task, but executed the trained turn rather than turning towards the visual cue during test, evidencing response learning. In Experiment 3, we investigated whether a compound visual cue could control salamanders' behaviour when it was the only cue predictive of reinforcement in a cross-maze by varying start position and cue placement. All learned to turn in the direction indicated by the compound visual cue, indicating that visual cues can come to control their behaviour. Following training, testing revealed that salamanders attended to stimuli foreground over background features. Overall, these results suggest that salamanders learn to execute responses over learning to use visual cues but can use visual cues if required. Our success with this paradigm offers the potential in future studies to explore salamanders' cognition further, as well as to shed light on how features of the tiger salamanders' life history (e.g. hibernation and metamorphosis) impact cognition.

Intelligent multiagent coordination based on reinforcement hierarchical neuro-fuzzy models.

PubMed

Mendoza, Leonardo Forero; Vellasco, Marley; Figueiredo, Karla

2014-12-01

This paper presents the research and development of two hybrid neuro-fuzzy models for the hierarchical coordination of multiple intelligent agents. The main objective of the models is to have multiple agents interact intelligently with each other in complex systems. We developed two new models of coordination for intelligent multiagent systems, which integrates the Reinforcement Learning Hierarchical Neuro-Fuzzy model with two proposed coordination mechanisms: the MultiAgent Reinforcement Learning Hierarchical Neuro-Fuzzy with a market-driven coordination mechanism (MA-RL-HNFP-MD) and the MultiAgent Reinforcement Learning Hierarchical Neuro-Fuzzy with graph coordination (MA-RL-HNFP-CG). In order to evaluate the proposed models and verify the contribution of the proposed coordination mechanisms, two multiagent benchmark applications were developed: the pursuit game and the robot soccer simulation. The results obtained demonstrated that the proposed coordination mechanisms greatly improve the performance of the multiagent system when compared with other strategies.

Refining Linear Fuzzy Rules by Reinforcement Learning

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.; Khedkar, Pratap S.; Malkani, Anil

1996-01-01

Linear fuzzy rules are increasingly being used in the development of fuzzy logic systems. Radial basis functions have also been used in the antecedents of the rules for clustering in product space which can automatically generate a set of linear fuzzy rules from an input/output data set. Manual methods are usually used in refining these rules. This paper presents a method for refining the parameters of these rules using reinforcement learning which can be applied in domains where supervised input-output data is not available and reinforcements are received only after a long sequence of actions. This is shown for a generalization of radial basis functions. The formation of fuzzy rules from data and their automatic refinement is an important step in closing the gap between the application of reinforcement learning methods in the domains where only some limited input-output data is available.

Instructed knowledge shapes feedback-driven aversive learning in striatum and orbitofrontal cortex, but not the amygdala

PubMed Central

Atlas, Lauren Y; Doll, Bradley B; Li, Jian; Daw, Nathaniel D; Phelps, Elizabeth A

2016-01-01

Socially-conveyed rules and instructions strongly shape expectations and emotions. Yet most neuroscientific studies of learning consider reinforcement history alone, irrespective of knowledge acquired through other means. We examined fear conditioning and reversal in humans to test whether instructed knowledge modulates the neural mechanisms of feedback-driven learning. One group was informed about contingencies and reversals. A second group learned only from reinforcement. We combined quantitative models with functional magnetic resonance imaging and found that instructions induced dissociations in the neural systems of aversive learning. Responses in striatum and orbitofrontal cortex updated with instructions and correlated with prefrontal responses to instructions. Amygdala responses were influenced by reinforcement similarly in both groups and did not update with instructions. Results extend work on instructed reward learning and reveal novel dissociations that have not been observed with punishments or rewards. Findings support theories of specialized threat-detection and may have implications for fear maintenance in anxiety. DOI: http://dx.doi.org/10.7554/eLife.15192.001 PMID:27171199

Applying the Explicit Time Central Difference Method for Numerical Simulation of the Dynamic Behavior of Elastoplastic Flexible Reinforced Plates

NASA Astrophysics Data System (ADS)

Yankovskii, A. P.

2017-12-01

Based on a stepwise algorithm involving central finite differences for the approximation in time, a mathematical model is developed for elastoplastic deformation of cross-reinforced plates with isotropically hardening materials of components of the composition. The model allows obtaining the solution of elastoplastic problems at discrete points in time by an explicit scheme. The initial boundary value problem of the dynamic behavior of flexible plates reinforced in their own plane is formulated in the von Kármán approximation with allowance for their weakened resistance to the transverse shear. With a common approach, the resolving equations corresponding to two variants of the Timoshenko theory are obtained. An explicit "cross" scheme for numerical integration of the posed initial boundary value problem has been constructed. The scheme is consistent with the incremental algorithm used for simulating the elastoplastic behavior of a reinforced medium. Calculations of the dynamic behavior have been performed for elastoplastic cylindrical bending of differently reinforced fiberglass rectangular elongated plates. It is shown that the reinforcement structure significantly affects their elastoplastic dynamic behavior. It has been found that the classical theory of plates is as a rule unacceptable for carrying out the required calculations (except for very thin plates), and the first version of the Timoshenko theory yields reasonable results only in cases of relatively thin constructions reinforced by lowmodulus fibers. Proceeding from the results of the work, it is recommended to use the second variant of the Timoshenko theory (as a more accurate one) for calculations of the elastoplastic behavior of reinforced plates.

Strength Analysis of Glass-Fiber-Reinforced Plastic during Buckling,

DTIC Science & Technology

An algorithm is developed for calculating and analyzing the stress tensor by the experimental function of deflections during the buckling of glass ... fiber -reinforced plastic shells loaded with a hydrostatic load. Malmeyster’s theory of strength is used to qualitatively establish the possible points of shell failure. (Author-PL)

Modeling Blast Loading on Buried Reinforced Concrete Structures with Zapotec

DOE PAGES

Bessette, Greg C.

2008-01-01

A coupled Euler-Lagrange solution approach is used to model the response of a buried reinforced concrete structure subjected to a close-in detonation of a high explosive charge. The coupling algorithm is discussed along with a set of benchmark calculations involving detonations in clay and sand.

Multi-agents and learning: Implications for Webusage mining.

PubMed

Lotfy, Hewayda M S; Khamis, Soheir M S; Aboghazalah, Maie M

2016-03-01

Characterization of user activities is an important issue in the design and maintenance of websites. Server weblog files have abundant information about the user's current interests. This information can be mined and analyzed therefore the administrators may be able to guide the users in their browsing activity so they may obtain relevant information in a shorter span of time to obtain user satisfaction. Web-based technology facilitates the creation of personally meaningful and socially useful knowledge through supportive interactions, communication and collaboration among educators, learners and information. This paper suggests a new methodology based on learning techniques for a Web-based Multiagent-based application to discover the hidden patterns in the user's visited links. It presents a new approach that involves unsupervised, reinforcement learning, and cooperation between agents. It is utilized to discover patterns that represent the user's profiles in a sample website into specific categories of materials using significance percentages. These profiles are used to make recommendations of interesting links and categories to the user. The experimental results of the approach showed successful user pattern recognition, and cooperative learning among agents to obtain user profiles. It indicates that combining different learning algorithms is capable of improving user satisfaction indicated by the percentage of precision, recall, the progressive category weight and F 1-measure.

Multi-agents and learning: Implications for Webusage mining

PubMed Central

Lotfy, Hewayda M.S.; Khamis, Soheir M.S.; Aboghazalah, Maie M.

2015-01-01

Characterization of user activities is an important issue in the design and maintenance of websites. Server weblog files have abundant information about the user’s current interests. This information can be mined and analyzed therefore the administrators may be able to guide the users in their browsing activity so they may obtain relevant information in a shorter span of time to obtain user satisfaction. Web-based technology facilitates the creation of personally meaningful and socially useful knowledge through supportive interactions, communication and collaboration among educators, learners and information. This paper suggests a new methodology based on learning techniques for a Web-based Multiagent-based application to discover the hidden patterns in the user’s visited links. It presents a new approach that involves unsupervised, reinforcement learning, and cooperation between agents. It is utilized to discover patterns that represent the user’s profiles in a sample website into specific categories of materials using significance percentages. These profiles are used to make recommendations of interesting links and categories to the user. The experimental results of the approach showed successful user pattern recognition, and cooperative learning among agents to obtain user profiles. It indicates that combining different learning algorithms is capable of improving user satisfaction indicated by the percentage of precision, recall, the progressive category weight and F1-measure. PMID:26966569

Life Span Differences in Electrophysiological Correlates of Monitoring Gains and Losses during Probabilistic Reinforcement Learning

ERIC Educational Resources Information Center

Hammerer, Dorothea; Li, Shu-Chen; Muller, Viktor; Lindenberger, Ulman

2011-01-01

By recording the feedback-related negativity (FRN) in response to gains and losses, we investigated the contribution of outcome monitoring mechanisms to age-associated differences in probabilistic reinforcement learning. Specifically, we assessed the difference of the monitoring reactions to gains and losses to investigate the monitoring of…

Reinforcement Learning in Young Adults with Developmental Language Impairment

ERIC Educational Resources Information Center

Lee, Joanna C.; Tomblin, J. Bruce

2012-01-01

The aim of the study was to examine reinforcement learning (RL) in young adults with developmental language impairment (DLI) within the context of a neurocomputational model of the basal ganglia-dopamine system (Frank, Seeberger, & O'Reilly, 2004). Two groups of young adults, one with DLI and the other without, were recruited. A probabilistic…

Effective Reinforcement Techniques in Elementary Physical Education: The Key to Behavior Management

ERIC Educational Resources Information Center

Downing, John; Keating, Tedd; Bennett, Carl

2005-01-01

The ability to shape appropriate behavior while extinguishing misbehavior is critical to teaching and learning in physical education. The scientific principles that affect student learning in the gymnasium also apply to the methods teachers use to influence social behaviors. Research indicates that reinforcement strategies are more effective than…

A neural model of hierarchical reinforcement learning

PubMed Central

Rasmussen, Daniel; Eliasmith, Chris

2017-01-01

We develop a novel, biologically detailed neural model of reinforcement learning (RL) processes in the brain. This model incorporates a broad range of biological features that pose challenges to neural RL, such as temporally extended action sequences, continuous environments involving unknown time delays, and noisy/imprecise computations. Most significantly, we expand the model into the realm of hierarchical reinforcement learning (HRL), which divides the RL process into a hierarchy of actions at different levels of abstraction. Here we implement all the major components of HRL in a neural model that captures a variety of known anatomical and physiological properties of the brain. We demonstrate the performance of the model in a range of different environments, in order to emphasize the aim of understanding the brain’s general reinforcement learning ability. These results show that the model compares well to previous modelling work and demonstrates improved performance as a result of its hierarchical ability. We also show that the model’s behaviour is consistent with available data on human hierarchical RL, and generate several novel predictions. PMID:28683111

Reinforcement Learning Trees

PubMed Central

Zhu, Ruoqing; Zeng, Donglin; Kosorok, Michael R.

2015-01-01

In this paper, we introduce a new type of tree-based method, reinforcement learning trees (RLT), which exhibits significantly improved performance over traditional methods such as random forests (Breiman, 2001) under high-dimensional settings. The innovations are three-fold. First, the new method implements reinforcement learning at each selection of a splitting variable during the tree construction processes. By splitting on the variable that brings the greatest future improvement in later splits, rather than choosing the one with largest marginal effect from the immediate split, the constructed tree utilizes the available samples in a more efficient way. Moreover, such an approach enables linear combination cuts at little extra computational cost. Second, we propose a variable muting procedure that progressively eliminates noise variables during the construction of each individual tree. The muting procedure also takes advantage of reinforcement learning and prevents noise variables from being considered in the search for splitting rules, so that towards terminal nodes, where the sample size is small, the splitting rules are still constructed from only strong variables. Last, we investigate asymptotic properties of the proposed method under basic assumptions and discuss rationale in general settings. PMID:26903687

Reinforcement learning state estimator.

PubMed

Morimoto, Jun; Doya, Kenji

2007-03-01

In this study, we propose a novel use of reinforcement learning for estimating hidden variables and parameters of nonlinear dynamical systems. A critical issue in hidden-state estimation is that we cannot directly observe estimation errors. However, by defining errors of observable variables as a delayed penalty, we can apply a reinforcement learning frame-work to state estimation problems. Specifically, we derive a method to construct a nonlinear state estimator by finding an appropriate feedback input gain using the policy gradient method. We tested the proposed method on single pendulum dynamics and show that the joint angle variable could be successfully estimated by observing only the angular velocity, and vice versa. In addition, we show that we could acquire a state estimator for the pendulum swing-up task in which a swing-up controller is also acquired by reinforcement learning simultaneously. Furthermore, we demonstrate that it is possible to estimate the dynamics of the pendulum itself while the hidden variables are estimated in the pendulum swing-up task. Application of the proposed method to a two-linked biped model is also presented.

Reciprocity Family Counseling: A Multi-Ethnic Model.

ERIC Educational Resources Information Center

Penrose, David M.

The Reciprocity Family Counseling Method involves learning principles of behavior modification including selective reinforcement, behavioral contracting, self-correction, and over-correction. Selective reinforcement refers to the recognition and modification of parent/child responses and reinforcers. Parents and children are asked to identify…

An intelligent agent for optimal river-reservoir system management

NASA Astrophysics Data System (ADS)

Rieker, Jeffrey D.; Labadie, John W.

2012-09-01

A generalized software package is presented for developing an intelligent agent for stochastic optimization of complex river-reservoir system management and operations. Reinforcement learning is an approach to artificial intelligence for developing a decision-making agent that learns the best operational policies without the need for explicit probabilistic models of hydrologic system behavior. The agent learns these strategies experientially in a Markov decision process through observational interaction with the environment and simulation of the river-reservoir system using well-calibrated models. The graphical user interface for the reinforcement learning process controller includes numerous learning method options and dynamic displays for visualizing the adaptive behavior of the agent. As a case study, the generalized reinforcement learning software is applied to developing an intelligent agent for optimal management of water stored in the Truckee river-reservoir system of California and Nevada for the purpose of streamflow augmentation for water quality enhancement. The intelligent agent successfully learns long-term reservoir operational policies that specifically focus on mitigating water temperature extremes during persistent drought periods that jeopardize the survival of threatened and endangered fish species.

Online Pedagogical Tutorial Tactics Optimization Using Genetic-Based Reinforcement Learning

PubMed Central

Lin, Hsuan-Ta; Lee, Po-Ming; Hsiao, Tzu-Chien

2015-01-01

Tutorial tactics are policies for an Intelligent Tutoring System (ITS) to decide the next action when there are multiple actions available. Recent research has demonstrated that when the learning contents were controlled so as to be the same, different tutorial tactics would make difference in students' learning gains. However, the Reinforcement Learning (RL) techniques that were used in previous studies to induce tutorial tactics are insufficient when encountering large problems and hence were used in offline manners. Therefore, we introduced a Genetic-Based Reinforcement Learning (GBML) approach to induce tutorial tactics in an online-learning manner without basing on any preexisting dataset. The introduced method can learn a set of rules from the environment in a manner similar to RL. It includes a genetic-based optimizer for rule discovery task by generating new rules from the old ones. This increases the scalability of a RL learner for larger problems. The results support our hypothesis about the capability of the GBML method to induce tutorial tactics. This suggests that the GBML method should be favorable in developing real-world ITS applications in the domain of tutorial tactics induction. PMID:26065018

Online Pedagogical Tutorial Tactics Optimization Using Genetic-Based Reinforcement Learning.

PubMed

Lin, Hsuan-Ta; Lee, Po-Ming; Hsiao, Tzu-Chien

2015-01-01

Tutorial tactics are policies for an Intelligent Tutoring System (ITS) to decide the next action when there are multiple actions available. Recent research has demonstrated that when the learning contents were controlled so as to be the same, different tutorial tactics would make difference in students' learning gains. However, the Reinforcement Learning (RL) techniques that were used in previous studies to induce tutorial tactics are insufficient when encountering large problems and hence were used in offline manners. Therefore, we introduced a Genetic-Based Reinforcement Learning (GBML) approach to induce tutorial tactics in an online-learning manner without basing on any preexisting dataset. The introduced method can learn a set of rules from the environment in a manner similar to RL. It includes a genetic-based optimizer for rule discovery task by generating new rules from the old ones. This increases the scalability of a RL learner for larger problems. The results support our hypothesis about the capability of the GBML method to induce tutorial tactics. This suggests that the GBML method should be favorable in developing real-world ITS applications in the domain of tutorial tactics induction.

A comparison of differential reinforcement procedures with children with autism.

PubMed

Boudreau, Brittany A; Vladescu, Jason C; Kodak, Tiffany M; Argott, Paul J; Kisamore, April N

2015-12-01

The current evaluation compared the effects of 2 differential reinforcement arrangements and a nondifferential reinforcement arrangement on the acquisition of tacts for 3 children with autism. Participants learned in all reinforcement-based conditions, and we discuss areas for future research in light of these findings and potential limitations. © Society for the Experimental Analysis of Behavior.

Mapping anhedonia onto reinforcement learning: a behavioural meta-analysis

PubMed Central

2013-01-01

Background Depression is characterised partly by blunted reactions to reward. However, tasks probing this deficiency have not distinguished insensitivity to reward from insensitivity to the prediction errors for reward that determine learning and are putatively reported by the phasic activity of dopamine neurons. We attempted to disentangle these factors with respect to anhedonia in the context of stress, Major Depressive Disorder (MDD), Bipolar Disorder (BPD) and a dopaminergic challenge. Methods Six behavioural datasets involving 392 experimental sessions were subjected to a model-based, Bayesian meta-analysis. Participants across all six studies performed a probabilistic reward task that used an asymmetric reinforcement schedule to assess reward learning. Healthy controls were tested under baseline conditions, stress or after receiving the dopamine D2 agonist pramipexole. In addition, participants with current or past MDD or BPD were evaluated. Reinforcement learning models isolated the contributions of variation in reward sensitivity and learning rate. Results MDD and anhedonia reduced reward sensitivity more than they affected the learning rate, while a low dose of the dopamine D2 agonist pramipexole showed the opposite pattern. Stress led to a pattern consistent with a mixed effect on reward sensitivity and learning rate. Conclusion Reward-related learning reflected at least two partially separable contributions. The first related to phasic prediction error signalling, and was preferentially modulated by a low dose of the dopamine agonist pramipexole. The second related directly to reward sensitivity, and was preferentially reduced in MDD and anhedonia. Stress altered both components. Collectively, these findings highlight the contribution of model-based reinforcement learning meta-analysis for dissecting anhedonic behavior. PMID:23782813

Reinforcement learning signals in the human striatum distinguish learners from nonlearners during reward-based decision making.

PubMed

Schönberg, Tom; Daw, Nathaniel D; Joel, Daphna; O'Doherty, John P

2007-11-21

The computational framework of reinforcement learning has been used to forward our understanding of the neural mechanisms underlying reward learning and decision-making behavior. It is known that humans vary widely in their performance in decision-making tasks. Here, we used a simple four-armed bandit task in which subjects are almost evenly split into two groups on the basis of their performance: those who do learn to favor choice of the optimal action and those who do not. Using models of reinforcement learning we sought to determine the neural basis of these intrinsic differences in performance by scanning both groups with functional magnetic resonance imaging. We scanned 29 subjects while they performed the reward-based decision-making task. Our results suggest that these two groups differ markedly in the degree to which reinforcement learning signals in the striatum are engaged during task performance. While the learners showed robust prediction error signals in both the ventral and dorsal striatum during learning, the nonlearner group showed a marked absence of such signals. Moreover, the magnitude of prediction error signals in a region of dorsal striatum correlated significantly with a measure of behavioral performance across all subjects. These findings support a crucial role of prediction error signals, likely originating from dopaminergic midbrain neurons, in enabling learning of action selection preferences on the basis of obtained rewards. Thus, spontaneously observed individual differences in decision making performance demonstrate the suggested dependence of this type of learning on the functional integrity of the dopaminergic striatal system in humans.

Reinforcement Learning Explains Conditional Cooperation and Its Moody Cousin.

PubMed

Ezaki, Takahiro; Horita, Yutaka; Takezawa, Masanori; Masuda, Naoki

2016-07-01

Direct reciprocity, or repeated interaction, is a main mechanism to sustain cooperation under social dilemmas involving two individuals. For larger groups and networks, which are probably more relevant to understanding and engineering our society, experiments employing repeated multiplayer social dilemma games have suggested that humans often show conditional cooperation behavior and its moody variant. Mechanisms underlying these behaviors largely remain unclear. Here we provide a proximate account for this behavior by showing that individuals adopting a type of reinforcement learning, called aspiration learning, phenomenologically behave as conditional cooperator. By definition, individuals are satisfied if and only if the obtained payoff is larger than a fixed aspiration level. They reinforce actions that have resulted in satisfactory outcomes and anti-reinforce those yielding unsatisfactory outcomes. The results obtained in the present study are general in that they explain extant experimental results obtained for both so-called moody and non-moody conditional cooperation, prisoner's dilemma and public goods games, and well-mixed groups and networks. Different from the previous theory, individuals are assumed to have no access to information about what other individuals are doing such that they cannot explicitly use conditional cooperation rules. In this sense, myopic aspiration learning in which the unconditional propensity of cooperation is modulated in every discrete time step explains conditional behavior of humans. Aspiration learners showing (moody) conditional cooperation obeyed a noisy GRIM-like strategy. This is different from the Pavlov, a reinforcement learning strategy promoting mutual cooperation in two-player situations.

Social stress reactivity alters reward and punishment learning

PubMed Central

Frank, Michael J.; Allen, John J. B.

2011-01-01

To examine how stress affects cognitive functioning, individual differences in trait vulnerability (punishment sensitivity) and state reactivity (negative affect) to social evaluative threat were examined during concurrent reinforcement learning. Lower trait-level punishment sensitivity predicted better reward learning and poorer punishment learning; the opposite pattern was found in more punishment sensitive individuals. Increasing state-level negative affect was directly related to punishment learning accuracy in highly punishment sensitive individuals, but these measures were inversely related in less sensitive individuals. Combined electrophysiological measurement, performance accuracy and computational estimations of learning parameters suggest that trait and state vulnerability to stress alter cortico-striatal functioning during reinforcement learning, possibly mediated via medio-frontal cortical systems. PMID:20453038

Social stress reactivity alters reward and punishment learning.

PubMed

Cavanagh, James F; Frank, Michael J; Allen, John J B

2011-06-01

To examine how stress affects cognitive functioning, individual differences in trait vulnerability (punishment sensitivity) and state reactivity (negative affect) to social evaluative threat were examined during concurrent reinforcement learning. Lower trait-level punishment sensitivity predicted better reward learning and poorer punishment learning; the opposite pattern was found in more punishment sensitive individuals. Increasing state-level negative affect was directly related to punishment learning accuracy in highly punishment sensitive individuals, but these measures were inversely related in less sensitive individuals. Combined electrophysiological measurement, performance accuracy and computational estimations of learning parameters suggest that trait and state vulnerability to stress alter cortico-striatal functioning during reinforcement learning, possibly mediated via medio-frontal cortical systems.

Predicting psychosis across diagnostic boundaries: Behavioral and computational modeling evidence for impaired reinforcement learning in schizophrenia and bipolar disorder with a history of psychosis.

PubMed

Strauss, Gregory P; Thaler, Nicholas S; Matveeva, Tatyana M; Vogel, Sally J; Sutton, Griffin P; Lee, Bern G; Allen, Daniel N

2015-08-01

There is increasing evidence that schizophrenia (SZ) and bipolar disorder (BD) share a number of cognitive, neurobiological, and genetic markers. Shared features may be most prevalent among SZ and BD with a history of psychosis. This study extended this literature by examining reinforcement learning (RL) performance in individuals with SZ (n = 29), BD with a history of psychosis (BD+; n = 24), BD without a history of psychosis (BD-; n = 23), and healthy controls (HC; n = 24). RL was assessed through a probabilistic stimulus selection task with acquisition and test phases. Computational modeling evaluated competing accounts of the data. Each participant's trial-by-trial decision-making behavior was fit to 3 computational models of RL: (a) a standard actor-critic model simulating pure basal ganglia-dependent learning, (b) a pure Q-learning model simulating action selection as a function of learned expected reward value, and (c) a hybrid model where an actor-critic is "augmented" by a Q-learning component, meant to capture the top-down influence of orbitofrontal cortex value representations on the striatum. The SZ group demonstrated greater reinforcement learning impairments at acquisition and test phases than the BD+, BD-, and HC groups. The BD+ and BD- groups displayed comparable performance at acquisition and test phases. Collapsing across diagnostic categories, greater severity of current psychosis was associated with poorer acquisition of the most rewarding stimuli as well as poor go/no-go learning at test. Model fits revealed that reinforcement learning in SZ was best characterized by a pure actor-critic model where learning is driven by prediction error signaling alone. In contrast, BD-, BD+, and HC were best fit by a hybrid model where prediction errors are influenced by top-down expected value representations that guide decision making. These findings suggest that abnormalities in the reward system are more prominent in SZ than BD; however, current psychotic symptoms may be associated with reinforcement learning deficits regardless of a Diagnostic and Statistical Manual of Mental Disorders (5th Edition; American Psychiatric Association, 2013) diagnosis. (c) 2015 APA, all rights reserved).

Motor Learning Enhances Use-Dependent Plasticity

PubMed Central

2017-01-01

Motor behaviors are shaped not only by current sensory signals but also by the history of recent experiences. For instance, repeated movements toward a particular target bias the subsequent movements toward that target direction. This process, called use-dependent plasticity (UDP), is considered a basic and goal-independent way of forming motor memories. Most studies consider movement history as the critical component that leads to UDP (Classen et al., 1998; Verstynen and Sabes, 2011). However, the effects of learning (i.e., improved performance) on UDP during movement repetition have not been investigated. Here, we used transcranial magnetic stimulation in two experiments to assess plasticity changes occurring in the primary motor cortex after individuals repeated reinforced and nonreinforced actions. The first experiment assessed whether learning a skill task modulates UDP. We found that a group that successfully learned the skill task showed greater UDP than a group that did not accumulate learning, but made comparable repeated actions. The second experiment aimed to understand the role of reinforcement learning in UDP while controlling for reward magnitude and action kinematics. We found that providing subjects with a binary reward without visual feedback of the cursor led to increased UDP effects. Subjects in the group that received comparable reward not associated with their actions maintained the previously induced UDP. Our findings illustrate how reinforcing consistent actions strengthens use-dependent memories and provide insight into operant mechanisms that modulate plastic changes in the motor cortex. SIGNIFICANCE STATEMENT Performing consistent motor actions induces use-dependent plastic changes in the motor cortex. This plasticity reflects one of the basic forms of human motor learning. Past studies assumed that this form of learning is exclusively affected by repetition of actions. However, here we showed that success-based reinforcement signals could affect the human use-dependent plasticity (UDP) process. Our results indicate that learning augments and interacts with UDP. This effect is important to the understanding of the interplay between the different forms of motor learning and suggests that reinforcement is not only important to learning new behaviors, but can shape our subsequent behavior via its interaction with UDP. PMID:28143961

An Advice Mechanism for Heterogeneous Robot Teams

NASA Astrophysics Data System (ADS)

Daniluk, Steven

The use of reinforcement learning for robot teams has enabled complex tasks to be performed, but at the cost of requiring a large amount of exploration. Exchanging information between robots in the form of advice is one method to accelerate performance improvements. This thesis presents an advice mechanism for robot teams that utilizes advice from heterogeneous advisers via a method guaranteeing convergence to an optimal policy. The presented mechanism has the capability to use multiple advisers at each time step, and decide when advice should be requested and accepted, such that the use of advice decreases over time. Additionally, collective collaborative, and cooperative behavioural algorithms are integrated into a robot team architecture, to create a new framework that provides fault tolerance and modularity for robot teams.

Decision theory, reinforcement learning, and the brain.

PubMed

Dayan, Peter; Daw, Nathaniel D

2008-12-01

Decision making is a core competence for animals and humans acting and surviving in environments they only partially comprehend, gaining rewards and punishments for their troubles. Decision-theoretic concepts permeate experiments and computational models in ethology, psychology, and neuroscience. Here, we review a well-known, coherent Bayesian approach to decision making, showing how it unifies issues in Markovian decision problems, signal detection psychophysics, sequential sampling, and optimal exploration and discuss paradigmatic psychological and neural examples of each problem. We discuss computational issues concerning what subjects know about their task and how ambitious they are in seeking optimal solutions; we address algorithmic topics concerning model-based and model-free methods for making choices; and we highlight key aspects of the neural implementation of decision making.

Specific effect of a dopamine partial agonist on counterfactual learning: evidence from Gilles de la Tourette syndrome.

PubMed

Salvador, Alexandre; Worbe, Yulia; Delorme, Cécile; Coricelli, Giorgio; Gaillard, Raphaël; Robbins, Trevor W; Hartmann, Andreas; Palminteri, Stefano

2017-07-24

The dopamine partial agonist aripiprazole is increasingly used to treat pathologies for which other antipsychotics are indicated because it displays fewer side effects, such as sedation and depression-like symptoms, than other dopamine receptor antagonists. Previously, we showed that aripiprazole may protect motivational function by preserving reinforcement-related signals used to sustain reward-maximization. However, the effect of aripiprazole on more cognitive facets of human reinforcement learning, such as learning from the forgone outcomes of alternative courses of action (i.e., counterfactual learning), is unknown. To test the influence of aripiprazole on counterfactual learning, we administered a reinforcement learning task that involves both direct learning from obtained outcomes and indirect learning from forgone outcomes to two groups of Gilles de la Tourette (GTS) patients, one consisting of patients who were completely unmedicated and the other consisting of patients who were receiving aripiprazole monotherapy, and to healthy subjects. We found that whereas learning performance improved in the presence of counterfactual feedback in both healthy controls and unmedicated GTS patients, this was not the case in aripiprazole-medicated GTS patients. Our results suggest that whereas aripiprazole preserves direct learning of action-outcome associations, it may impair more complex inferential processes, such as counterfactual learning from forgone outcomes, in GTS patients treated with this medication.

Somatosensory Contribution to the Initial Stages of Human Motor Learning

PubMed Central

Bernardi, Nicolò F.; Darainy, Mohammad

2015-01-01

The early stages of motor skill acquisition are often marked by uncertainty about the sensory and motor goals of the task, as is the case in learning to speak or learning the feel of a good tennis serve. Here we present an experimental model of this early learning process, in which targets are acquired by exploration and reinforcement rather than sensory error. We use this model to investigate the relative contribution of motor and sensory factors to human motor learning. Participants make active reaching movements or matched passive movements to an unseen target using a robot arm. We find that learning through passive movements paired with reinforcement is comparable with learning associated with active movement, both in terms of magnitude and durability, with improvements due to training still observable at a 1 week retest. Motor learning is also accompanied by changes in somatosensory perceptual acuity. No stable changes in motor performance are observed for participants that train, actively or passively, in the absence of reinforcement, or for participants who are given explicit information about target position in the absence of somatosensory experience. These findings indicate that the somatosensory system dominates learning in the early stages of motor skill acquisition. SIGNIFICANCE STATEMENT The research focuses on the initial stages of human motor learning, introducing a new experimental model that closely approximates the key features of motor learning outside of the laboratory. The finding indicates that it is the somatosensory system rather than the motor system that dominates learning in the early stages of motor skill acquisition. This is important given that most of our computational models of motor learning are based on the idea that learning is motoric in origin. This is also a valuable finding for rehabilitation of patients with limited mobility as it shows that reinforcement in conjunction with passive movement results in benefits to motor learning that are as great as those observed for active movement training. PMID:26490869

Somatic and Reinforcement-Based Plasticity in the Initial Stages of Human Motor Learning.

PubMed

Sidarta, Ananda; Vahdat, Shahabeddin; Bernardi, Nicolò F; Ostry, David J

2016-11-16

As one learns to dance or play tennis, the desired somatosensory state is typically unknown. Trial and error is important as motor behavior is shaped by successful and unsuccessful movements. As an experimental model, we designed a task in which human participants make reaching movements to a hidden target and receive positive reinforcement when successful. We identified somatic and reinforcement-based sources of plasticity on the basis of changes in functional connectivity using resting-state fMRI before and after learning. The neuroimaging data revealed reinforcement-related changes in both motor and somatosensory brain areas in which a strengthening of connectivity was related to the amount of positive reinforcement during learning. Areas of prefrontal cortex were similarly altered in relation to reinforcement, with connectivity between sensorimotor areas of putamen and the reward-related ventromedial prefrontal cortex strengthened in relation to the amount of successful feedback received. In other analyses, we assessed connectivity related to changes in movement direction between trials, a type of variability that presumably reflects exploratory strategies during learning. We found that connectivity in a network linking motor and somatosensory cortices increased with trial-to-trial changes in direction. Connectivity varied as well with the change in movement direction following incorrect movements. Here the changes were observed in a somatic memory and decision making network involving ventrolateral prefrontal cortex and second somatosensory cortex. Our results point to the idea that the initial stages of motor learning are not wholly motor but rather involve plasticity in somatic and prefrontal networks related both to reward and exploration. In the initial stages of motor learning, the placement of the limbs is learned primarily through trial and error. In an experimental analog, participants make reaching movements to a hidden target and receive positive feedback when successful. We identified sources of plasticity based on changes in functional connectivity using resting-state fMRI. The main finding is that there is a strengthening of connectivity between reward-related prefrontal areas and sensorimotor areas in the basal ganglia and frontal cortex. There is also a strengthening of connectivity related to movement exploration in sensorimotor circuits involved in somatic memory and decision making. The results indicate that initial stages of motor learning depend on plasticity in somatic and prefrontal networks related to reward and exploration. Copyright © 2016 the authors 0270-6474/16/3611682-11$15.00/0.

Somatic and Reinforcement-Based Plasticity in the Initial Stages of Human Motor Learning

PubMed Central

Sidarta, Ananda; Vahdat, Shahabeddin; Bernardi, Nicolò F.

2016-01-01

As one learns to dance or play tennis, the desired somatosensory state is typically unknown. Trial and error is important as motor behavior is shaped by successful and unsuccessful movements. As an experimental model, we designed a task in which human participants make reaching movements to a hidden target and receive positive reinforcement when successful. We identified somatic and reinforcement-based sources of plasticity on the basis of changes in functional connectivity using resting-state fMRI before and after learning. The neuroimaging data revealed reinforcement-related changes in both motor and somatosensory brain areas in which a strengthening of connectivity was related to the amount of positive reinforcement during learning. Areas of prefrontal cortex were similarly altered in relation to reinforcement, with connectivity between sensorimotor areas of putamen and the reward-related ventromedial prefrontal cortex strengthened in relation to the amount of successful feedback received. In other analyses, we assessed connectivity related to changes in movement direction between trials, a type of variability that presumably reflects exploratory strategies during learning. We found that connectivity in a network linking motor and somatosensory cortices increased with trial-to-trial changes in direction. Connectivity varied as well with the change in movement direction following incorrect movements. Here the changes were observed in a somatic memory and decision making network involving ventrolateral prefrontal cortex and second somatosensory cortex. Our results point to the idea that the initial stages of motor learning are not wholly motor but rather involve plasticity in somatic and prefrontal networks related both to reward and exploration. SIGNIFICANCE STATEMENT In the initial stages of motor learning, the placement of the limbs is learned primarily through trial and error. In an experimental analog, participants make reaching movements to a hidden target and receive positive feedback when successful. We identified sources of plasticity based on changes in functional connectivity using resting-state fMRI. The main finding is that there is a strengthening of connectivity between reward-related prefrontal areas and sensorimotor areas in the basal ganglia and frontal cortex. There is also a strengthening of connectivity related to movement exploration in sensorimotor circuits involved in somatic memory and decision making. The results indicate that initial stages of motor learning depend on plasticity in somatic and prefrontal networks related to reward and exploration. PMID:27852776

Adaptive Critic-based Neurofuzzy Controller for the Steam Generator Water Level

NASA Astrophysics Data System (ADS)

Fakhrazari, Amin; Boroushaki, Mehrdad

2008-06-01

In this paper, an adaptive critic-based neurofuzzy controller is presented for water level regulation of nuclear steam generators. The problem has been of great concern for many years as the steam generator is a highly nonlinear system showing inverse response dynamics especially at low operating power levels. Fuzzy critic-based learning is a reinforcement learning method based on dynamic programming. The only information available for the critic agent is the system feedback which is interpreted as the last action the controller has performed in the previous state. The signal produced by the critic agent is used alongside the backpropagation of error algorithm to tune online conclusion parts of the fuzzy inference rules. The critic agent here has a proportional-derivative structure and the fuzzy rule base has nine rules. The proposed controller shows satisfactory transient responses, disturbance rejection and robustness to model uncertainty. Its simple design procedure and structure, nominates it as one of the suitable controller designs for the steam generator water level control in nuclear power plant industry.

A Cognitively Grounded Measure of Pronunciation Distance

PubMed Central

Wieling, Martijn; Nerbonne, John; Bloem, Jelke; Gooskens, Charlotte; Heeringa, Wilbert; Baayen, R. Harald

2014-01-01

In this study we develop pronunciation distances based on naive discriminative learning (NDL). Measures of pronunciation distance are used in several subfields of linguistics, including psycholinguistics, dialectology and typology. In contrast to the commonly used Levenshtein algorithm, NDL is grounded in cognitive theory of competitive reinforcement learning and is able to generate asymmetrical pronunciation distances. In a first study, we validated the NDL-based pronunciation distances by comparing them to a large set of native-likeness ratings given by native American English speakers when presented with accented English speech. In a second study, the NDL-based pronunciation distances were validated on the basis of perceptual dialect distances of Norwegian speakers. Results indicated that the NDL-based pronunciation distances matched perceptual distances reasonably well with correlations ranging between 0.7 and 0.8. While the correlations were comparable to those obtained using the Levenshtein distance, the NDL-based approach is more flexible as it is also able to incorporate acoustic information other than sound segments. PMID:24416119

What is the optimal task difficulty for reinforcement learning of brain self-regulation?

PubMed

Bauer, Robert; Vukelić, Mathias; Gharabaghi, Alireza

2016-09-01

The balance between action and reward during neurofeedback may influence reinforcement learning of brain self-regulation. Eleven healthy volunteers participated in three runs of motor imagery-based brain-machine interface feedback where a robot passively opened the hand contingent to β-band modulation. For each run, the β-desynchronization threshold to initiate the hand robot movement increased in difficulty (low, moderate, and demanding). In this context, the incentive to learn was estimated by the change of reward per action, operationalized as the change in reward duration per movement onset. Variance analysis revealed a significant interaction between threshold difficulty and the relationship between reward duration and number of movement onsets (p<0.001), indicating a negative learning incentive for low difficulty, but a positive learning incentive for moderate and demanding runs. Exploration of different thresholds in the same data set indicated that the learning incentive peaked at higher thresholds than the threshold which resulted in maximum classification accuracy. Specificity is more important than sensitivity of neurofeedback for reinforcement learning of brain self-regulation. Learning efficiency requires adequate challenge by neurofeedback interventions. Copyright © 2016 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

Coexistence of Reward and Unsupervised Learning During the Operant Conditioning of Neural Firing Rates

PubMed Central

Kerr, Robert R.; Grayden, David B.; Thomas, Doreen A.; Gilson, Matthieu; Burkitt, Anthony N.

2014-01-01

A fundamental goal of neuroscience is to understand how cognitive processes, such as operant conditioning, are performed by the brain. Typical and well studied examples of operant conditioning, in which the firing rates of individual cortical neurons in monkeys are increased using rewards, provide an opportunity for insight into this. Studies of reward-modulated spike-timing-dependent plasticity (RSTDP), and of other models such as R-max, have reproduced this learning behavior, but they have assumed that no unsupervised learning is present (i.e., no learning occurs without, or independent of, rewards). We show that these models cannot elicit firing rate reinforcement while exhibiting both reward learning and ongoing, stable unsupervised learning. To fix this issue, we propose a new RSTDP model of synaptic plasticity based upon the observed effects that dopamine has on long-term potentiation and depression (LTP and LTD). We show, both analytically and through simulations, that our new model can exhibit unsupervised learning and lead to firing rate reinforcement. This requires that the strengthening of LTP by the reward signal is greater than the strengthening of LTD and that the reinforced neuron exhibits irregular firing. We show the robustness of our findings to spike-timing correlations, to the synaptic weight dependence that is assumed, and to changes in the mean reward. We also consider our model in the differential reinforcement of two nearby neurons. Our model aligns more strongly with experimental studies than previous models and makes testable predictions for future experiments. PMID:24475240

Oculomotor learning revisited: a model of reinforcement learning in the basal ganglia incorporating an efference copy of motor actions

PubMed Central

Fee, Michale S.

2012-01-01

In its simplest formulation, reinforcement learning is based on the idea that if an action taken in a particular context is followed by a favorable outcome, then, in the same context, the tendency to produce that action should be strengthened, or reinforced. While reinforcement learning forms the basis of many current theories of basal ganglia (BG) function, these models do not incorporate distinct computational roles for signals that convey context, and those that convey what action an animal takes. Recent experiments in the songbird suggest that vocal-related BG circuitry receives two functionally distinct excitatory inputs. One input is from a cortical region that carries context information about the current “time” in the motor sequence. The other is an efference copy of motor commands from a separate cortical brain region that generates vocal variability during learning. Based on these findings, I propose here a general model of vertebrate BG function that combines context information with a distinct motor efference copy signal. The signals are integrated by a learning rule in which efference copy inputs gate the potentiation of context inputs (but not efference copy inputs) onto medium spiny neurons in response to a rewarded action. The hypothesis is described in terms of a circuit that implements the learning of visually guided saccades. The model makes testable predictions about the anatomical and functional properties of hypothesized context and efference copy inputs to the striatum from both thalamic and cortical sources. PMID:22754501

Oculomotor learning revisited: a model of reinforcement learning in the basal ganglia incorporating an efference copy of motor actions.

PubMed

Fee, Michale S

2012-01-01

In its simplest formulation, reinforcement learning is based on the idea that if an action taken in a particular context is followed by a favorable outcome, then, in the same context, the tendency to produce that action should be strengthened, or reinforced. While reinforcement learning forms the basis of many current theories of basal ganglia (BG) function, these models do not incorporate distinct computational roles for signals that convey context, and those that convey what action an animal takes. Recent experiments in the songbird suggest that vocal-related BG circuitry receives two functionally distinct excitatory inputs. One input is from a cortical region that carries context information about the current "time" in the motor sequence. The other is an efference copy of motor commands from a separate cortical brain region that generates vocal variability during learning. Based on these findings, I propose here a general model of vertebrate BG function that combines context information with a distinct motor efference copy signal. The signals are integrated by a learning rule in which efference copy inputs gate the potentiation of context inputs (but not efference copy inputs) onto medium spiny neurons in response to a rewarded action. The hypothesis is described in terms of a circuit that implements the learning of visually guided saccades. The model makes testable predictions about the anatomical and functional properties of hypothesized context and efference copy inputs to the striatum from both thalamic and cortical sources.

Histidine-decarboxylase knockout mice show deficient nonreinforced episodic object memory, improved negatively reinforced water-maze performance, and increased neo- and ventro-striatal dopamine turnover.

PubMed

Dere, Ekrem; De Souza-Silva, Maria A; Topic, Bianca; Spieler, Richard E; Haas, Helmut L; Huston, Joseph P

2003-01-01

The brain's histaminergic system has been implicated in hippocampal synaptic plasticity, learning, and memory, as well as brain reward and reinforcement. Our past pharmacological and lesion studies indicated that the brain's histamine system exerts inhibitory effects on the brain's reinforcement respective reward system reciprocal to mesolimbic dopamine systems, thereby modulating learning and memory performance. Given the close functional relationship between brain reinforcement and memory processes, the total disruption of brain histamine synthesis via genetic disruption of its synthesizing enzyme, histidine decarboxylase (HDC), in the mouse might have differential effects on learning dependent on the task-inherent reinforcement contingencies. Here, we investigated the effects of an HDC gene disruption in the mouse in a nonreinforced object exploration task and a negatively reinforced water-maze task as well as on neo- and ventro-striatal dopamine systems known to be involved in brain reward and reinforcement. Histidine decarboxylase knockout (HDC-KO) mice had higher dihydrophenylacetic acid concentrations and a higher dihydrophenylacetic acid/dopamine ratio in the neostriatum. In the ventral striatum, dihydrophenylacetic acid/dopamine and 3-methoxytyramine/dopamine ratios were higher in HDC-KO mice. Furthermore, the HDC-KO mice showed improved water-maze performance during both hidden and cued platform tasks, but deficient object discrimination based on temporal relationships. Our data imply that disruption of brain histamine synthesis can have both memory promoting and suppressive effects via distinct and independent mechanisms and further indicate that these opposed effects are related to the task-inherent reinforcement contingencies.

Dorsal Striatal-Midbrain Connectivity in Humans Predicts How Reinforcements Are Used to Guide Decisions

ERIC Educational Resources Information Center

Kahnt, Thorsten; Park, Soyoung Q.; Cohen, Michael X.; Beck, Anne; Heinz, Andreas; Wrase, Jana

2009-01-01

It has been suggested that the target areas of dopaminergic midbrain neurons, the dorsal (DS) and ventral striatum (VS), are differently involved in reinforcement learning especially as actor and critic. Whereas the critic learns to predict rewards, the actor maintains action values to guide future decisions. The different midbrain connections to…

Autonomous Inter-Task Transfer in Reinforcement Learning Domains

DTIC Science & Technology

2008-08-01

Twentieth International Joint Conference on Artificial Intelli - gence, 2007. 304 Fumihide Tanaka and Masayuki Yamamura. Multitask reinforcement learning...Functions . . . . . . . . . . . . . . . . . . . . . . 17 2.2.3 Artificial Neural Networks . . . . . . . . . . . . . . . . . . . . 18 2.2.4 Instance-based...tures [Laird et al., 1986, Choi et al., 2007]. However, TL for RL tasks has only recently been gaining attention in the artificial intelligence

A look at Behaviourism and Perceptual Control Theory in Interface Design

DTIC Science & Technology

1998-02-01

behaviours such as response variability, instinctive drift, autoshaping , etc. Perceptual Control Theory (PCT) postulates that behaviours result from the...internal variables. Behaviourism, on the other hand, can not account for variability in responses, instinctive drift, autoshaping , etc. Researchers... Autoshaping . Animals appear to learn without reinforcement. However, conditioning theory speculates that learning results only when reinforcement

BEHAVIORAL MECHANISMS UNDERLYING NICOTINE REINFORCEMENT

PubMed Central

Rupprecht, Laura E.; Smith, Tracy T.; Schassburger, Rachel L.; Buffalari, Deanne M.; Sved, Alan F.; Donny, Eric C.

2015-01-01

Cigarette smoking is the leading cause of preventable deaths worldwide and nicotine, the primary psychoactive constituent in tobacco, drives sustained use. The behavioral actions of nicotine are complex and extend well beyond the actions of the drug as a primary reinforcer. Stimuli that are consistently paired with nicotine can, through associative learning, take on reinforcing properties as conditioned stimuli. These conditioned stimuli can then impact the rate and probability of behavior and even function as conditioning reinforcers that maintain behavior in the absence of nicotine. Nicotine can also act as a conditioned stimulus, predicting the delivery of other reinforcers, which may allow nicotine to acquire value as a conditioned reinforcer. These associative effects, establishing non-nicotine stimuli as conditioned stimuli with discriminative stimulus and conditioned reinforcing properties as well as establishing nicotine as a conditioned stimulus, are predicted by basic conditioning principles. However, nicotine can also act non-associatively. Nicotine directly enhances the reinforcing efficacy of other reinforcing stimuli in the environment, an effect that does not require a temporal or predictive relationship between nicotine and either the stimulus or the behavior. Hence, the reinforcing actions of nicotine stem both from the primary reinforcing actions of the drug (and the subsequent associative learning effects) as well as the reinforcement enhancement action of nicotine which is non-associative in nature. Gaining a better understanding of how nicotine impacts behavior will allow for maximally effective tobacco control efforts aimed at reducing the harm associated with tobacco use by reducing and/or treating its addictiveness. PMID:25638333

Measuring reinforcement learning and motivation constructs in experimental animals: relevance to the negative symptoms of schizophrenia.

PubMed

Markou, Athina; Salamone, John D; Bussey, Timothy J; Mar, Adam C; Brunner, Daniela; Gilmour, Gary; Balsam, Peter

2013-11-01

The present review article summarizes and expands upon the discussions that were initiated during a meeting of the Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia (CNTRICS; http://cntrics.ucdavis.edu) meeting. A major goal of the CNTRICS meeting was to identify experimental procedures and measures that can be used in laboratory animals to assess psychological constructs that are related to the psychopathology of schizophrenia. The issues discussed in this review reflect the deliberations of the Motivation Working Group of the CNTRICS meeting, which included most of the authors of this article as well as additional participants. After receiving task nominations from the general research community, this working group was asked to identify experimental procedures in laboratory animals that can assess aspects of reinforcement learning and motivation that may be relevant for research on the negative symptoms of schizophrenia, as well as other disorders characterized by deficits in reinforcement learning and motivation. The tasks described here that assess reinforcement learning are the Autoshaping Task, Probabilistic Reward Learning Tasks, and the Response Bias Probabilistic Reward Task. The tasks described here that assess motivation are Outcome Devaluation and Contingency Degradation Tasks and Effort-Based Tasks. In addition to describing such methods and procedures, the present article provides a working vocabulary for research and theory in this field, as well as an industry perspective about how such tasks may be used in drug discovery. It is hoped that this review can aid investigators who are conducting research in this complex area, promote translational studies by highlighting shared research goals and fostering a common vocabulary across basic and clinical fields, and facilitate the development of medications for the treatment of symptoms mediated by reinforcement learning and motivational deficits. Copyright © 2013 Elsevier Ltd. All rights reserved.

Goal-Directed and Habit-Like Modulations of Stimulus Processing during Reinforcement Learning.

PubMed

Luque, David; Beesley, Tom; Morris, Richard W; Jack, Bradley N; Griffiths, Oren; Whitford, Thomas J; Le Pelley, Mike E

2017-03-15

Recent research has shown that perceptual processing of stimuli previously associated with high-value rewards is automatically prioritized even when rewards are no longer available. It has been hypothesized that such reward-related modulation of stimulus salience is conceptually similar to an "attentional habit." Recording event-related potentials in humans during a reinforcement learning task, we show strong evidence in favor of this hypothesis. Resistance to outcome devaluation (the defining feature of a habit) was shown by the stimulus-locked P1 component, reflecting activity in the extrastriate visual cortex. Analysis at longer latencies revealed a positive component (corresponding to the P3b, from 550-700 ms) sensitive to outcome devaluation. Therefore, distinct spatiotemporal patterns of brain activity were observed corresponding to habitual and goal-directed processes. These results demonstrate that reinforcement learning engages both attentional habits and goal-directed processes in parallel. Consequences for brain and computational models of reinforcement learning are discussed. SIGNIFICANCE STATEMENT The human attentional network adapts to detect stimuli that predict important rewards. A recent hypothesis suggests that the visual cortex automatically prioritizes reward-related stimuli, driven by cached representations of reward value; that is, stimulus-response habits. Alternatively, the neural system may track the current value of the predicted outcome. Our results demonstrate for the first time that visual cortex activity is increased for reward-related stimuli even when the rewarding event is temporarily devalued. In contrast, longer-latency brain activity was specifically sensitive to transient changes in reward value. Therefore, we show that both habit-like attention and goal-directed processes occur in the same learning episode at different latencies. This result has important consequences for computational models of reinforcement learning. Copyright © 2017 the authors 0270-6474/17/373009-09$15.00/0.

Measuring reinforcement learning and motivation constructs in experimental animals: relevance to the negative symptoms of schizophrenia

PubMed Central

Markou, Athina; Salamone, John D.; Bussey, Timothy; Mar, Adam; Brunner, Daniela; Gilmour, Gary; Balsam, Peter

2013-01-01

The present review article summarizes and expands upon the discussions that were initiated during a meeting of the Cognitive Neuroscience Treatment Research to Improve Cognition in Schizophrenia (CNTRICS; http://cntrics.ucdavis.edu). A major goal of the CNTRICS meeting was to identify experimental procedures and measures that can be used in laboratory animals to assess psychological constructs that are related to the psychopathology of schizophrenia. The issues discussed in this review reflect the deliberations of the Motivation Working Group of the CNTRICS meeting, which included most of the authors of this article as well as additional participants. After receiving task nominations from the general research community, this working group was asked to identify experimental procedures in laboratory animals that can assess aspects of reinforcement learning and motivation that may be relevant for research on the negative symptoms of schizophrenia, as well as other disorders characterized by deficits in reinforcement learning and motivation. The tasks described here that assess reinforcement learning are the Autoshaping Task, Probabilistic Reward Learning Tasks, and the Response Bias Probabilistic Reward Task. The tasks described here that assess motivation are Outcome Devaluation and Contingency Degradation Tasks and Effort-Based Tasks. In addition to describing such methods and procedures, the present article provides a working vocabulary for research and theory in this field, as well as an industry perspective about how such tasks may be used in drug discovery. It is hoped that this review can aid investigators who are conducting research in this complex area, promote translational studies by highlighting shared research goals and fostering a common vocabulary across basic and clinical fields, and facilitate the development of medications for the treatment of symptoms mediated by reinforcement learning and motivational deficits. PMID:23994273

The role of within-compound associations in learning about absent cues.

PubMed

Witnauer, James E; Miller, Ralph R

2011-05-01

When two cues are reinforced together (in compound), most associative models assume that animals learn an associative network that includes direct cue-outcome associations and a within-compound association. All models of associative learning subscribe to the importance of cue-outcome associations, but most models assume that within-compound associations are irrelevant to each cue's subsequent behavioral control. In the present article, we present an extension of Van Hamme and Wasserman's (Learning and Motivation 25:127-151, 1994) model of retrospective revaluation based on learning about absent cues that are retrieved through within-compound associations. The model was compared with a model lacking retrieval through within-compound associations. Simulations showed that within-compound associations are necessary for the model to explain higher-order retrospective revaluation and the observed greater retrospective revaluation after partial reinforcement than after continuous reinforcement alone. These simulations suggest that the associability of an absent stimulus is determined by the extent to which the stimulus is activated through the within-compound association.

The role of within-compound associations in learning about absent cues

PubMed Central

Witnauer, James E.

2011-01-01

When two cues are reinforced together (in compound), most associative models assume that animals learn an associative network that includes direct cue–outcome associations and a within-compound association. All models of associative learning subscribe to the importance of cue–outcome associations, but most models assume that within-compound associations are irrelevant to each cue's subsequent behavioral control. In the present article, we present an extension of Van Hamme and Wasserman's (Learning and Motivation 25:127–151, 1994) model of retrospective revaluation based on learning about absent cues that are retrieved through within-compound associations. The model was compared with a model lacking retrieval through within-compound associations. Simulations showed that within-compound associations are necessary for the model to explain higher-order retrospective revaluation and the observed greater retrospective revaluation after partial reinforcement than after continuous reinforcement alone. These simulations suggest that the associability of an absent stimulus is determined by the extent to which the stimulus is activated through the within-compound association. PMID:21264569

Pleasurable music affects reinforcement learning according to the listener

PubMed Central

Gold, Benjamin P.; Frank, Michael J.; Bogert, Brigitte; Brattico, Elvira

2013-01-01

Mounting evidence links the enjoyment of music to brain areas implicated in emotion and the dopaminergic reward system. In particular, dopamine release in the ventral striatum seems to play a major role in the rewarding aspect of music listening. Striatal dopamine also influences reinforcement learning, such that subjects with greater dopamine efficacy learn better to approach rewards while those with lesser dopamine efficacy learn better to avoid punishments. In this study, we explored the practical implications of musical pleasure through its ability to facilitate reinforcement learning via non-pharmacological dopamine elicitation. Subjects from a wide variety of musical backgrounds chose a pleasurable and a neutral piece of music from an experimenter-compiled database, and then listened to one or both of these pieces (according to pseudo-random group assignment) as they performed a reinforcement learning task dependent on dopamine transmission. We assessed musical backgrounds as well as typical listening patterns with the new Helsinki Inventory of Music and Affective Behaviors (HIMAB), and separately investigated behavior for the training and test phases of the learning task. Subjects with more musical experience trained better with neutral music and tested better with pleasurable music, while those with less musical experience exhibited the opposite effect. HIMAB results regarding listening behaviors and subjective music ratings indicate that these effects arose from different listening styles: namely, more affective listening in non-musicians and more analytical listening in musicians. In conclusion, musical pleasure was able to influence task performance, and the shape of this effect depended on group and individual factors. These findings have implications in affective neuroscience, neuroaesthetics, learning, and music therapy. PMID:23970875

Network Supervision of Adult Experience and Learning Dependent Sensory Cortical Plasticity.

PubMed

Blake, David T

2017-06-18

The brain is capable of remodeling throughout life. The sensory cortices provide a useful preparation for studying neuroplasticity both during development and thereafter. In adulthood, sensory cortices change in the cortical area activated by behaviorally relevant stimuli, by the strength of response within that activated area, and by the temporal profiles of those responses. Evidence supports forms of unsupervised, reinforcement, and fully supervised network learning rules. Studies on experience-dependent plasticity have mostly not controlled for learning, and they find support for unsupervised learning mechanisms. Changes occur with greatest ease in neurons containing α-CamKII, which are pyramidal neurons in layers II/III and layers V/VI. These changes use synaptic mechanisms including long term depression. Synaptic strengthening at NMDA-containing synapses does occur, but its weak association with activity suggests other factors also initiate changes. Studies that control learning find support of reinforcement learning rules and limited evidence of other forms of supervised learning. Behaviorally associating a stimulus with reinforcement leads to a strengthening of cortical response strength and enlarging of response area with poor selectivity. Associating a stimulus with omission of reinforcement leads to a selective weakening of responses. In some preparations in which these associations are not as clearly made, neurons with the most informative discharges are relatively stronger after training. Studies analyzing the temporal profile of responses associated with omission of reward, or of plasticity in studies with different discriminanda but statistically matched stimuli, support the existence of limited supervised network learning. © 2017 American Physiological Society. Compr Physiol 7:977-1008, 2017. Copyright © 2017 John Wiley & Sons, Inc.

Feedback from the heart: Emotional learning and memory is controlled by cardiac cycle, interoceptive accuracy and personality.

PubMed

Pfeifer, Gaby; Garfinkel, Sarah N; Gould van Praag, Cassandra D; Sahota, Kuljit; Betka, Sophie; Critchley, Hugo D

2017-05-01

Feedback processing is critical to trial-and-error learning. Here, we examined whether interoceptive signals concerning the state of cardiovascular arousal influence the processing of reinforcing feedback during the learning of 'emotional' face-name pairs, with subsequent effects on retrieval. Participants (N=29) engaged in a learning task of face-name pairs (fearful, neutral, happy faces). Correct and incorrect learning decisions were reinforced by auditory feedback, which was delivered either at cardiac systole (on the heartbeat, when baroreceptors signal the contraction of the heart to the brain), or at diastole (between heartbeats during baroreceptor quiescence). We discovered a cardiac influence on feedback processing that enhanced the learning of fearful faces in people with heightened interoceptive ability. Individuals with enhanced accuracy on a heartbeat counting task learned fearful face-name pairs better when feedback was given at systole than at diastole. This effect was not present for neutral and happy faces. At retrieval, we also observed related effects of personality: First, individuals scoring higher for extraversion showed poorer retrieval accuracy. These individuals additionally manifested lower resting heart rate and lower state anxiety, suggesting that attenuated levels of cardiovascular arousal in extraverts underlies poorer performance. Second, higher extraversion scores predicted higher emotional intensity ratings of fearful faces reinforced at systole. Third, individuals scoring higher for neuroticism showed higher retrieval confidence for fearful faces reinforced at diastole. Our results show that cardiac signals shape feedback processing to influence learning of fearful faces, an effect underpinned by personality differences linked to psychophysiological arousal. Copyright © 2017 Elsevier B.V. All rights reserved.

A reinforcement learning-based architecture for fuzzy logic control

NASA Technical Reports Server (NTRS)

Berenji, Hamid R.

1992-01-01

This paper introduces a new method for learning to refine a rule-based fuzzy logic controller. A reinforcement learning technique is used in conjunction with a multilayer neural network model of a fuzzy controller. The approximate reasoning based intelligent control (ARIC) architecture proposed here learns by updating its prediction of the physical system's behavior and fine tunes a control knowledge base. Its theory is related to Sutton's temporal difference (TD) method. Because ARIC has the advantage of using the control knowledge of an experienced operator and fine tuning it through the process of learning, it learns faster than systems that train networks from scratch. The approach is applied to a cart-pole balancing system.

A cholinergic feedback circuit to regulate striatal population uncertainty and optimize reinforcement learning.

PubMed

Franklin, Nicholas T; Frank, Michael J

2015-12-25

Convergent evidence suggests that the basal ganglia support reinforcement learning by adjusting action values according to reward prediction errors. However, adaptive behavior in stochastic environments requires the consideration of uncertainty to dynamically adjust the learning rate. We consider how cholinergic tonically active interneurons (TANs) may endow the striatum with such a mechanism in computational models spanning three Marr's levels of analysis. In the neural model, TANs modulate the excitability of spiny neurons, their population response to reinforcement, and hence the effective learning rate. Long TAN pauses facilitated robustness to spurious outcomes by increasing divergence in synaptic weights between neurons coding for alternative action values, whereas short TAN pauses facilitated stochastic behavior but increased responsiveness to change-points in outcome contingencies. A feedback control system allowed TAN pauses to be dynamically modulated by uncertainty across the spiny neuron population, allowing the system to self-tune and optimize performance across stochastic environments.

Online human training of a myoelectric prosthesis controller via actor-critic reinforcement learning.

PubMed

Pilarski, Patrick M; Dawson, Michael R; Degris, Thomas; Fahimi, Farbod; Carey, Jason P; Sutton, Richard S

2011-01-01

As a contribution toward the goal of adaptable, intelligent artificial limbs, this work introduces a continuous actor-critic reinforcement learning method for optimizing the control of multi-function myoelectric devices. Using a simulated upper-arm robotic prosthesis, we demonstrate how it is possible to derive successful limb controllers from myoelectric data using only a sparse human-delivered training signal, without requiring detailed knowledge about the task domain. This reinforcement-based machine learning framework is well suited for use by both patients and clinical staff, and may be easily adapted to different application domains and the needs of individual amputees. To our knowledge, this is the first my-oelectric control approach that facilitates the online learning of new amputee-specific motions based only on a one-dimensional (scalar) feedback signal provided by the user of the prosthesis. © 2011 IEEE

Cardiac Concomitants of Feedback and Prediction Error Processing in Reinforcement Learning.

PubMed

Kastner, Lucas; Kube, Jana; Villringer, Arno; Neumann, Jane

2017-01-01

Successful learning hinges on the evaluation of positive and negative feedback. We assessed differential learning from reward and punishment in a monetary reinforcement learning paradigm, together with cardiac concomitants of positive and negative feedback processing. On the behavioral level, learning from reward resulted in more advantageous behavior than learning from punishment, suggesting a differential impact of reward and punishment on successful feedback-based learning. On the autonomic level, learning and feedback processing were closely mirrored by phasic cardiac responses on a trial-by-trial basis: (1) Negative feedback was accompanied by faster and prolonged heart rate deceleration compared to positive feedback. (2) Cardiac responses shifted from feedback presentation at the beginning of learning to stimulus presentation later on. (3) Most importantly, the strength of phasic cardiac responses to the presentation of feedback correlated with the strength of prediction error signals that alert the learner to the necessity for behavioral adaptation. Considering participants' weight status and gender revealed obesity-related deficits in learning to avoid negative consequences and less consistent behavioral adaptation in women compared to men. In sum, our results provide strong new evidence for the notion that during learning phasic cardiac responses reflect an internal value and feedback monitoring system that is sensitive to the violation of performance-based expectations. Moreover, inter-individual differences in weight status and gender may affect both behavioral and autonomic responses in reinforcement-based learning.

Cardiac Concomitants of Feedback and Prediction Error Processing in Reinforcement Learning

PubMed Central

Kastner, Lucas; Kube, Jana; Villringer, Arno; Neumann, Jane

2017-01-01

Successful learning hinges on the evaluation of positive and negative feedback. We assessed differential learning from reward and punishment in a monetary reinforcement learning paradigm, together with cardiac concomitants of positive and negative feedback processing. On the behavioral level, learning from reward resulted in more advantageous behavior than learning from punishment, suggesting a differential impact of reward and punishment on successful feedback-based learning. On the autonomic level, learning and feedback processing were closely mirrored by phasic cardiac responses on a trial-by-trial basis: (1) Negative feedback was accompanied by faster and prolonged heart rate deceleration compared to positive feedback. (2) Cardiac responses shifted from feedback presentation at the beginning of learning to stimulus presentation later on. (3) Most importantly, the strength of phasic cardiac responses to the presentation of feedback correlated with the strength of prediction error signals that alert the learner to the necessity for behavioral adaptation. Considering participants' weight status and gender revealed obesity-related deficits in learning to avoid negative consequences and less consistent behavioral adaptation in women compared to men. In sum, our results provide strong new evidence for the notion that during learning phasic cardiac responses reflect an internal value and feedback monitoring system that is sensitive to the violation of performance-based expectations. Moreover, inter-individual differences in weight status and gender may affect both behavioral and autonomic responses in reinforcement-based learning. PMID:29163004

Learning the specific quality of taste reinforcement in larval Drosophila

PubMed Central

Schleyer, Michael; Miura, Daisuke; Tanimura, Teiichi; Gerber, Bertram

2015-01-01

The only property of reinforcement insects are commonly thought to learn about is its value. We show that larval Drosophila not only remember the value of reinforcement (How much?), but also its quality (What?). This is demonstrated both within the appetitive domain by using sugar vs amino acid as different reward qualities, and within the aversive domain by using bitter vs high-concentration salt as different qualities of punishment. From the available literature, such nuanced memories for the quality of reinforcement are unexpected and pose a challenge to present models of how insect memory is organized. Given that animals as simple as larval Drosophila, endowed with but 10,000 neurons, operate with both reinforcement value and quality, we suggest that both are fundamental aspects of mnemonic processing—in any brain. DOI: http://dx.doi.org/10.7554/eLife.04711.001 PMID:25622533

Evidence for a neural law of effect.

PubMed

Athalye, Vivek R; Santos, Fernando J; Carmena, Jose M; Costa, Rui M

2018-03-02

Thorndike's law of effect states that actions that lead to reinforcements tend to be repeated more often. Accordingly, neural activity patterns leading to reinforcement are also reentered more frequently. Reinforcement relies on dopaminergic activity in the ventral tegmental area (VTA), and animals shape their behavior to receive dopaminergic stimulation. Seeking evidence for a neural law of effect, we found that mice learn to reenter more frequently motor cortical activity patterns that trigger optogenetic VTA self-stimulation. Learning was accompanied by gradual shaping of these patterns, with participating neurons progressively increasing and aligning their covariance to that of the target pattern. Motor cortex patterns that lead to phasic dopaminergic VTA activity are progressively reinforced and shaped, suggesting a mechanism by which animals select and shape actions to reliably achieve reinforcement. Copyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works.

Active-learning strategies: the use of a game to reinforce learning in nursing education. A case study.

PubMed

Boctor, Lisa

2013-03-01

The majority of nursing students are kinesthetic learners, preferring a hands-on, active approach to education. Research shows that active-learning strategies can increase student learning and satisfaction. This study looks at the use of one active-learning strategy, a Jeopardy-style game, 'Nursopardy', to reinforce Fundamentals of Nursing material, aiding in students' preparation for a standardized final exam. The game was created keeping students varied learning styles and the NCLEX blueprint in mind. The blueprint was used to create 5 categories, with 26 total questions. Student survey results, using a five-point Likert scale showed that they did find this learning method enjoyable and beneficial to learning. More research is recommended regarding learning outcomes, when using active-learning strategies, such as games. Copyright © 2012 Elsevier Ltd. All rights reserved.

The "proactive" model of learning: Integrative framework for model-free and model-based reinforcement learning utilizing the associative learning-based proactive brain concept.

PubMed

Zsuga, Judit; Biro, Klara; Papp, Csaba; Tajti, Gabor; Gesztelyi, Rudolf

2016-02-01

Reinforcement learning (RL) is a powerful concept underlying forms of associative learning governed by the use of a scalar reward signal, with learning taking place if expectations are violated. RL may be assessed using model-based and model-free approaches. Model-based reinforcement learning involves the amygdala, the hippocampus, and the orbitofrontal cortex (OFC). The model-free system involves the pedunculopontine-tegmental nucleus (PPTgN), the ventral tegmental area (VTA) and the ventral striatum (VS). Based on the functional connectivity of VS, model-free and model based RL systems center on the VS that by integrating model-free signals (received as reward prediction error) and model-based reward related input computes value. Using the concept of reinforcement learning agent we propose that the VS serves as the value function component of the RL agent. Regarding the model utilized for model-based computations we turned to the proactive brain concept, which offers an ubiquitous function for the default network based on its great functional overlap with contextual associative areas. Hence, by means of the default network the brain continuously organizes its environment into context frames enabling the formulation of analogy-based association that are turned into predictions of what to expect. The OFC integrates reward-related information into context frames upon computing reward expectation by compiling stimulus-reward and context-reward information offered by the amygdala and hippocampus, respectively. Furthermore we suggest that the integration of model-based expectations regarding reward into the value signal is further supported by the efferent of the OFC that reach structures canonical for model-free learning (e.g., the PPTgN, VTA, and VS). (c) 2016 APA, all rights reserved).

Incorporating Dispositional Traits into the Treatment of Anorexia Nervosa

PubMed Central

Herzog, David; Moskovich, Ashley; Merwin, Rhonda; Lin, Tammy

2014-01-01

We provide a general framework to guide the development of interventions that aim to address persistent features in eating disorders that may preclude effective treatment. Using perfectionism as an exemplar, we draw from research in cognitive neuroscience regarding attention and reinforcement learning, from learning theory and social psychology regarding vicarious learning and implications for the role modeling of significant others, and from clinical psychology on the importance of verbal narratives as barriers that may influence expectations and shape reinforcement schedules. PMID:21243482

Hybrid learning in signalling games

NASA Astrophysics Data System (ADS)

Barrett, Jeffrey A.; Cochran, Calvin T.; Huttegger, Simon; Fujiwara, Naoki

2017-09-01

Lewis-Skyrms signalling games have been studied under a variety of low-rationality learning dynamics. Reinforcement dynamics are stable but slow and prone to evolving suboptimal signalling conventions. A low-inertia trial-and-error dynamical like win-stay/lose-randomise is fast and reliable at finding perfect signalling conventions but unstable in the context of noise or agent error. Here we consider a low-rationality hybrid of reinforcement and win-stay/lose-randomise learning that exhibits the virtues of both. This hybrid dynamics is reliable, stable and exceptionally fast.

Discrete Serotonin Systems Mediate Memory Enhancement and Escape Latencies after Unpredicted Aversive Experience in Drosophila Place Memory

PubMed Central

Sitaraman, Divya; Kramer, Elizabeth F.; Kahsai, Lily; Ostrowski, Daniela; Zars, Troy

2017-01-01

Feedback mechanisms in operant learning are critical for animals to increase reward or reduce punishment. However, not all conditions have a behavior that can readily resolve an event. Animals must then try out different behaviors to better their situation through outcome learning. This form of learning allows for novel solutions and with positive experience can lead to unexpected behavioral routines. Learned helplessness, as a type of outcome learning, manifests in part as increases in escape latency in the face of repeated unpredicted shocks. Little is known about the mechanisms of outcome learning. When fruit fly Drosophila melanogaster are exposed to unpredicted high temperatures in a place learning paradigm, flies both increase escape latencies and have a higher memory when given control of a place/temperature contingency. Here we describe discrete serotonin neuronal circuits that mediate aversive reinforcement, escape latencies, and memory levels after place learning in the presence and absence of unexpected aversive events. The results show that two features of learned helplessness depend on the same modulatory system as aversive reinforcement. Moreover, changes in aversive reinforcement and escape latency depend on local neural circuit modulation, while memory enhancement requires larger modulation of multiple behavioral control circuits. PMID:29321732

Efficient Grammar Induction Algorithm with Parse Forests from Real Corpora

NASA Astrophysics Data System (ADS)

Kurihara, Kenichi; Kameya, Yoshitaka; Sato, Taisuke

The task of inducing grammar structures has received a great deal of attention. The reasons why researchers have studied are different; to use grammar induction as the first stage in building large treebanks or to make up better language models. However, grammar induction has inherent computational complexity. To overcome it, some grammar induction algorithms add new production rules incrementally. They refine the grammar while keeping their computational complexity low. In this paper, we propose a new efficient grammar induction algorithm. Although our algorithm is similar to algorithms which learn a grammar incrementally, our algorithm uses the graphical EM algorithm instead of the Inside-Outside algorithm. We report results of learning experiments in terms of learning speeds. The results show that our algorithm learns a grammar in constant time regardless of the size of the grammar. Since our algorithm decreases syntactic ambiguities in each step, our algorithm reduces required time for learning. This constant-time learning considerably affects learning time for larger grammars. We also reports results of evaluation of criteria to choose nonterminals. Our algorithm refines a grammar based on a nonterminal in each step. Since there can be several criteria to decide which nonterminal is the best, we evaluate them by learning experiments.

Tonic or Phasic Stimulation of Dopaminergic Projections to Prefrontal Cortex Causes Mice to Maintain or Deviate from Previously Learned Behavioral Strategies

PubMed Central

Ellwood, Ian T.; Patel, Tosha; Wadia, Varun; Lee, Anthony T.; Liptak, Alayna T.

2017-01-01

Dopamine neurons in the ventral tegmental area (VTA) encode reward prediction errors and can drive reinforcement learning through their projections to striatum, but much less is known about their projections to prefrontal cortex (PFC). Here, we studied these projections and observed phasic VTA–PFC fiber photometry signals after the delivery of rewards. Next, we studied how optogenetic stimulation of these projections affects behavior using conditioned place preference and a task in which mice learn associations between cues and food rewards and then use those associations to make choices. Neither phasic nor tonic stimulation of dopaminergic VTA–PFC projections elicited place preference. Furthermore, substituting phasic VTA–PFC stimulation for food rewards was not sufficient to reinforce new cue–reward associations nor maintain previously learned ones. However, the same patterns of stimulation that failed to reinforce place preference or cue–reward associations were able to modify behavior in other ways. First, continuous tonic stimulation maintained previously learned cue–reward associations even after they ceased being valid. Second, delivering phasic stimulation either continuously or after choices not previously associated with reward induced mice to make choices that deviated from previously learned associations. In summary, despite the fact that dopaminergic VTA–PFC projections exhibit phasic increases in activity that are time locked to the delivery of rewards, phasic activation of these projections does not necessarily reinforce specific actions. Rather, dopaminergic VTA–PFC activity can control whether mice maintain or deviate from previously learned cue–reward associations. SIGNIFICANCE STATEMENT Dopaminergic inputs from ventral tegmental area (VTA) to striatum encode reward prediction errors and reinforce specific actions; however, it is currently unknown whether dopaminergic inputs to prefrontal cortex (PFC) play similar or distinct roles. Here, we used bulk Ca2+ imaging to show that unexpected rewards or reward-predicting cues elicit phasic increases in the activity of dopaminergic VTA–PFC fibers. However, in multiple behavioral paradigms, we failed to observe reinforcing effects after stimulation of these fibers. In these same experiments, we did find that tonic or phasic patterns of stimulation caused mice to maintain or deviate from previously learned cue–reward associations, respectively. Therefore, although they may exhibit similar patterns of activity, dopaminergic inputs to striatum and PFC can elicit divergent behavioral effects. PMID:28739583

The Effect of a Token Reinforcement Program on the Reading Comprehension of a Learning Disabled Student.

ERIC Educational Resources Information Center

Galbreath, Joy; Feldman, David

The relationship of reading comprehension accuracy and a contingently administered token reinforcement program used with an elementary level learning disabled student in the classroom was examined. The S earned points for each correct answer made after oral reading sessions. At the conclusion of the class he could exchange his points for rewards.…

Cocaine addiction as a homeostatic reinforcement learning disorder.

PubMed

Keramati, Mehdi; Durand, Audrey; Girardeau, Paul; Gutkin, Boris; Ahmed, Serge H

2017-03-01

Drug addiction implicates both reward learning and homeostatic regulation mechanisms of the brain. This has stimulated 2 partially successful theoretical perspectives on addiction. Many important aspects of addiction, however, remain to be explained within a single, unified framework that integrates the 2 mechanisms. Building upon a recently developed homeostatic reinforcement learning theory, the authors focus on a key transition stage of addiction that is well modeled in animals, escalation of drug use, and propose a computational theory of cocaine addiction where cocaine reinforces behavior due to its rapid homeostatic corrective effect, whereas its chronic use induces slow and long-lasting changes in homeostatic setpoint. Simulations show that our new theory accounts for key behavioral and neurobiological features of addiction, most notably, escalation of cocaine use, drug-primed craving and relapse, individual differences underlying dose-response curves, and dopamine D2-receptor downregulation in addicts. The theory also generates unique predictions about cocaine self-administration behavior in rats that are confirmed by new experimental results. Viewing addiction as a homeostatic reinforcement learning disorder coherently explains many behavioral and neurobiological aspects of the transition to cocaine addiction, and suggests a new perspective toward understanding addiction. (PsycINFO Database Record (c) 2017 APA, all rights reserved).