Filters








123,117 Hits in 4.0 sec

Persistent Rule-based Interactive Reinforcement Learning [article]

Adam Bignold and Francisco Cruz and Richard Dazeley and Peter Vamplew and Cameron Foale
2021 arXiv   pre-print
In this work, we propose a persistent rule-based interactive reinforcement learning approach, i.e., a method for retaining and reusing provided knowledge, allowing trainers to give general advice relevant  ...  Moreover, rule-based advice shows similar performance impact as state-based advice, but with a substantially reduced interaction count.  ...  In this regard, rules allow advice to be provided that generalises over multiple states. 2 Rule-based Interactive Advice Reinforcement Learning and Interactive Advice Reinforcement learning (RL) [9]  ... 
arXiv:2102.02441v2 fatcat:ds6myvafkbbt3c3vu5nh47ttei

Page 732 of Psychological Abstracts Vol. 69, Issue 3 [page]

1983 Psychological Abstracts  
Findings have implications for teachers who, based on the subtleness and flexibility of the rules for behaving found in the present research, may be more influential in developing sex-rules among adolescents  ...  . —16 preschool children were identified as having an internal or external locus of control based on the Stephens-Delys Reinforcement Contingency Interview and were observed inter- acting with peers, teachers  ... 

Multi-agent Learning and the Reinforcement Gradient [chapter]

Michael Kaisers, Daan Bloembergen, Karl Tuyls
2012 Lecture Notes in Computer Science  
This article tackles the diversification by showing a persistent principle in several independent reinforcement learning algorithms that have been applied to multi-agent settings.  ...  The number of proposed reinforcement learning algorithms appears to be ever-growing.  ...  However, also value based learning is based on the same dynamics although the learning update rules appear to be very different. Q-learning does not update the policy vector directly.  ... 
doi:10.1007/978-3-642-34799-3_10 fatcat:j447aprslnafphio2bvyr3f3ya

Modeling Competitive Strategy Choice and Cognitive Learning Dynamics

Yeonjoo Min, Hani S. Mahmassani
2007 Transportation Research Record  
Players can learn how to play those games over time (game) and can update their belief by learning how to play based on their past experience and acquired knowledge of the opponents' actions and their  ...  , dynamic nature of competitive auction-based marketplaces.  ...  Average payoff, cumulative choice sequence, and habit persistence variables are used in the behavioral reinforcement learning model specification in Tables 2 and 3 .  ... 
doi:10.3141/2032-01 fatcat:h4swkwgfabazno4araaxekeouu

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

Elien Segers, Tom Beckers, Hilde Geurts, Laurence Claes, Marina Danckaerts, Saskia van der Oord
2018 Frontiers in Psychology  
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  ...  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),  ...  reinforced children learned the response-sequence rule faster during acquisition and showed faster extinction of this previously learned rule during extinction than partially reinforced children (Figure  ... 
doi:10.3389/fpsyg.2018.00394 pmid:29643822 pmcid:PMC5882844 fatcat:k5gm7gwrf5c6lfdgr4xz6p2q7m

Learning social behavior

Maja J. Matarić
1997 Robotics and Autonomous Systems  
We describe three sources of reinforcement and show their effectiveness in learning nongreedy social rules.  ...  Lnteractions, in this paper we describe how, given the substrate, greedy agents can learn social rules that benefit the group ',as a whole.  ...  Acknowledgements The research reported here was done partly at the MIT Artificial Intelligence Laboratory, and subsequently continued at the author's Interaction Laboratory at Brandeis University, in the  ... 
doi:10.1016/s0921-8890(96)00068-1 fatcat:d4a3vpw4znfzhcf26s3ghgr5le

Uncertainty-Dependent Extinction of Fear Memory in an Amygdala-mPFC Neural Circuit Model

Yuzhe Li, Ken Nakae, Shin Ishii, Honda Naoki, Samuel J. Gershman
2016 PLoS Computational Biology  
Here, we developed a neural circuit model based on three distinct types of neurons (fear, persistent and extinction neurons) in the amygdala and medial prefrontal cortex (mPFC).  ...  To adapt to uncertain situations, they flexibly learn to associate environmental cues with rewards and punishments.  ...  In eq (4), the synaptic weight of the CS input to the fear neural unit, w F , is modulated, according to the Rescorla-Wagner learning rule [43, 44] , based on the prediction error of the net severity.  ... 
doi:10.1371/journal.pcbi.1005099 pmid:27617747 pmcid:PMC5019407 fatcat:ym3ihpzzkvfmxjj5zf5a3llefa

Self-Adapting Payoff Matrices in Repeated Interactions

Siang Chong, Xin Yao
2006 2006 IEEE Symposium on Computational Intelligence and Games  
on the matrix based on reinforcement feedback from game interactions payoff matrix [I I].  ...  in turn affects the learning of strategy actions because they restrict strategies from adapting (e.g., behaviors. learning) their individual payoff matrices based on feedback  ...  of strategy behaviors can be learned through a process of adaptation based on for future interactions.  ... 
doi:10.1109/cig.2006.311688 dblp:conf/cig/ChongY06 fatcat:cnlvjbsvszdxblxnczkrgvq6fm

Motivated reinforcement learning for non-player characters in persistent computer game worlds

Kathryn Merrick, Mary Lou Maher
2006 Proceedings of the 2006 ACM SIGCHI international conference on Advances in computer entertainment technology - ACE '06  
This paper presents motivated reinforcement learning agents as a means of creating non-player characters that can both evolve and adapt.  ...  Motivated reinforcement learning agents explore their environment and learn new behaviours in response to interesting experiences, allowing them to display progressively evolving behavioural patterns.  ...  Such interactions with human players are usually facilitated through a graphical or text based user interface.  ... 
doi:10.1145/1178823.1178828 dblp:conf/ACMace/MerrickM06 fatcat:usp4pptyzbegnnjgjnsykt6eoi

Carle's Game: An Open-Ended Challenge in Exploratory Machine Creativity [article]

Q. Tyrell Davis
2021 arXiv   pre-print
Carle's Game is based on machine agent interaction with CARLE, a Cellular Automata Reinforcement Learning Environment.  ...  It is an introduction to CARLE, a Life-like cellular automata simulator and reinforcement learning environment.  ...  Starter agents include Cellular Automata Reinforcement Learning Agent (CARLA), a policy based on continuous-valued cellular automata rules, and HARLI (Hebbian Automata Reinforcement Learning Improviser  ... 
arXiv:2107.05786v1 fatcat:v4qu7gyhirhkhnqhafg4n2e7p4

Rule-governed behavior: An ongoing RFT-based operant analysis

Heloisa Ribeiro Zapparoli, Ramon Marin, Colin Harte
2021 Perspectivas em Análise do Comportamento  
the complexities involved in rule-following in the face of competing reinforcement contingencies, a phenomenon typically linked to human psychological suffering.  ...  To this end, we provide an overview of an RFT-based operant account of rule-following as it currently stands, and outline a recent program of experimental research that has utilized this approach to explore  ...  As we will later argue, it is this very interaction that may be of particular importance in coming to understand the complexities involved in persistent rule-following in the face of competing reinforcement  ... 
doi:10.18761/pac.2021.v12.rft.09 fatcat:oe26x5bmsbb2rk6ubfwq2ubflm

Ventrolateral prefrontal cortex is required for performance of a strategy implementation task but not reinforcer devaluation effects in rhesus monkeys

Mark G. Baxter, David Gaffan, Diana A. Kyriazis, Anna S. Mitchell
2009 European Journal of Neuroscience  
Monkeys with ventrolateral prefrontal lesions were impaired in performing the strategy-based task, but not on value-based decision-making.  ...  The ability to apply behavioral strategies to obtain rewards efficiently and make choices based on changes in the value of rewards is fundamental to the adaptive control of behavior.  ...  A focused contrast in the overall anova evaluated the three-way interaction of number of persistent choices, pre ⁄ post, and lesion group, based on the hypothesis that strategy implementation deteriorates  ... 
doi:10.1111/j.1460-9568.2009.06740.x pmid:19453635 pmcid:PMC2688497 fatcat:tapn2ahy7rgctldpflvjx6ek5q

Efficient Value Function Approximation with Unsupervised Hierarchical Categorization for a Reinforcement Learning Agent

Yongjia Wang, John E. Laird
2010 2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology  
We also explore the types of prior domain knowledge that can be used to regulate learning based on the characteristics of environment.  ...  We investigate the problem of reinforcement learning (RL) in a challenging object-oriented environment, where the functional diversity of objects is high, and the agent must learn quickly by generalizing  ...  Our approach is a unique synthesis of two machine learning approaches: unsupervised, hierarchical category learning and behavior adaptation based on reward signals using reinforcement learning (RL) [1  ... 
doi:10.1109/wi-iat.2010.16 dblp:conf/iat/WangL10 fatcat:c6dduyyulfftlkcalplkpsrnaa

Behavior-analytic approaches to decision making

Edmund Fantino
2004 Behavioural Processes  
The interaction of the rule with the novel contingencies and both the historical and contemporary context in which the interaction occurs, largely determines whether the rule is applied fruitfully or misapplied  ...  Indeed our interpretation of sunk-cost phenomena (when they occur) is that lessons (or "rules") learned about persistence are eagerly applied (and misapplied!).  ... 
doi:10.1016/j.beproc.2004.03.009 pmid:15157977 fatcat:l3xzvuy4pffyddstv5e47ep2cy

Behavior-analytic approaches to decision making

E FANTINO
2004 Behavioural Processes  
The interaction of the rule with the novel contingencies and both the historical and contemporary context in which the interaction occurs, largely determines whether the rule is applied fruitfully or misapplied  ...  Indeed our interpretation of sunk-cost phenomena (when they occur) is that lessons (or "rules") learned about persistence are eagerly applied (and misapplied!).  ... 
doi:10.1016/s0376-6357(04)00058-0 fatcat:sr5b4ei5ufedrcw2ebke76524y
« Previous Showing results 1 — 15 out of 123,117 results