665,392 Hits in 3.5 sec

Observational Learning by Reinforcement Learning [article]

Diana Borsa, Bilal Piot, Rémi Munos, Olivier Pietquin
2017 arXiv   pre-print
Especially, we argue that observational learning can emerge from pure Reinforcement Learning (RL), potentially coupled with memory.  ...  The later is naturally modeled by RL, by correlating the learning agent's reward with the teacher agent's behaviour.  ...  It demonstrates that observational learning can emerge from reinforcement learning, combined with memory.  ... 
arXiv:1706.06617v1 fatcat:373blxc2rnfqvksiskyhcqezoy

Learning a Transferable World Model by Reinforcement Agent in Deterministic Observable Grid-World Environments

Jurgita Kapočiūtė-Dzikienė, Gailius Raškinis
2012 Information Technology and Control  
The proposed agent was compared to the Q-learning and Adaptive Dynamic Programming based agents and demonstrated better ability to achieve goals in static observable deterministic gridworld environments  ...  Reinforcement-based agents have difficulties in transferring their acquired knowledge into new different environments due to the common identities-based percept representation and the lack of appropriate  ...  It combines Q-learning and supervised learning by learning Q-function with relational regression tree algorithm [4] .  ... 
doi:10.5755/j01.itc.41.4.915 fatcat:5pgeedmyjbhdzgmidzvkyneh4i

Reward Prediction Error Signaling during Reinforcement Learning in Social Anxiety Disorder is altered by Social Observation [article]

Michael P. I. Becker, Rolf Voegler, Jutta Peterburs, Christian Bellebaum, David Hofmann, Thomas Straube
2019 bioRxiv   pre-print
Reinforcement learning (RL) models offer a framework to define changes in outcome expectations in a formal way by computing the prediction error (PE).  ...  RESULTS: Coupling to prediction errors in SAD was elevated in dorsomedial prefrontal cortex (DMPFC) when learning under observation.  ...  CC-BY-NC-ND 4.0 International license certified by peer review) is the author/funder.  ... 
doi:10.1101/821512 fatcat:uraajs4nxjb65m3apwlr63uara

Space-Air-Ground Integrated Mobile Crowdsensing for Partially Observable Data Collection by Multi-Scale Convolutional Graph Reinforcement Learning

Yixiang Ren, Zhenhui Ye, Guanghua Song, Xiaohong Jiang
2022 Entropy  
To address this multi-agent problem, we propose a novel deep reinforcement learning (DRL) based method called Multi-Scale Soft Deep Recurrent Graph Network (ms-SDRGN).  ...  In addition, a stochastic policy can be learned through a maximum-entropy method with an adjustable temperature parameter.  ...  [35] applied Graph Convolutional Reinforcement Learning (DGN [36] ) in MBS.  ... 
doi:10.3390/e24050638 fatcat:ol64sxrv65cshaw3ucrtc23l64

Model-free Control of Partially Observable Underactuated Systems by pairing Reinforcement Learning with Delay Embeddings

Martinius Knudsen, Sverre Hendseth, Gunnar Tufte, Axel Sandvig
2022 Modeling, Identification and Control  
If partial observability can be overcome, these conditions suggest the use of reinforcement learning (RL).  ...  Instead, RL learns to map any observation to appropriate action (determined by a reward function), even if these observations do not lie on the original geometric state manifold.  ...  Reinforcement learning (RL) is a great candidate for this.  ... 
doi:10.4173/mic.2022.1.1 fatcat:djyqwv5ylfe3valbupnkhrxz3a

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

Jessica Neu
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 independent variables were the observation of reinforcement for correct responses as the control condition and the observation of corrections for incorrect responses.  ...  In addition to the expansion of communities of reinforcers by conditioning vocal praise for younger students via observation, research on learning academic skills via observation has been conducted with  ... 
doi:10.7916/d8wq0b1x fatcat:o6qp3v756zgfbfa3eb3cm7rl2e

Unsupervised re-scoring of observation probability in viterbi based on reinforcement learning by using confidence measure and HMM neighborhood

Carlos Molina, Nestor Becerra Yoma, Fernando Huenupán, Claudio Garreton
2007 Interspeech 2007   unpublished
A two-step Viterbi decoding based on reinforcement learning is described. The idea is to strength or weaken HMM's by using Bayes-based confidence measure (BBCM) and distances between models.  ...  Correction component based on reinforcement learning The motivation is to apply the reinforcement learning principle to correct or re-estimate observation probabilities [4] .  ...  As was mentioned above, reinforcement learning could be applied as a correction in the observation probabilities.  ... 
doi:10.21437/interspeech.2007-486 fatcat:vkexbogfcnbdvmzfqa7uve7uni

A Social View to Multiagent Reinforcement Learning

Qiang WEI, Tetsuo SAWARAGI
2004 Transactions of the Society of Instrument and Control Engineers  
Through a modeling of four software robot agents' cooperative work to balance a ball on a plate, two social RL approaches-observing reinforcement and vicarious reinforcementare applied to individual RL  ...  This paper presents a study on Multiagent Reinforcement Learning (RL) for cooperating work from a social view to solve the problems of individual agent's incomplete world model, conflict of individual  ...  Acknowledgments This paper is supported in part by Center of Excellence for Research and Education on Complex Functional Mechanical Systems (COE program of the Ministry of Education, Culture, Sports, Science  ... 
doi:10.9746/sicetr1965.40.328 fatcat:uf4mxseeejf7jnkzmsebxvbbta

Observational Learning from a Radical-Behavioristic Viewpoint

Hikaru Deguchi
1984 The Behavior Analyst  
cognitive mediation, and (3) vicarious reinforcement.  ...  Finally, the origin of observational learning is discussed in terms of recent data of neonatal imitation.  ...  The definition may stand only until we observe some longer-term reinforcement control of observational learning.  ... 
doi:10.1007/bf03391892 pmid:22478602 pmcid:PMC2741741 fatcat:xioac3lf4fgobnpp36rg3awtrq

Teacher's Reinforcement in Teaching English at High School Level

Dian Frahesti, Harum Natasha
2020 Lingua Susastra  
In data collection, researchers conducted interviews and observations on teachers of SMA Islam Negeri 2 Kuantan Singingi to determine the components of reinforcement used by teachers in learning English  ...  The reinforcement most often used by teachers in teaching English was reinforcement with words, while the reinforcement that was rarely used by teachers was token reinforcement.  ...  They stated that they often received token reinforcement in the learning process. To strengthen the results of the research, the researcher also collected the data by observation.  ... 
doi:10.24036/ls.v1i1.4 fatcat:my6qgqga4va57cqlsn3ztbn4qi

Page 140 of Viewpoints in Teaching and Learning Vol. 46, Issue 5 [page]

1970 Viewpoints in Teaching and Learning  
Observational learning occurs when we learn by watching another person without being directly reinforced for learning and probably without any overt practice.  ...  First, let us be clear what we mean by observational learning.  ... 

Hierarchical extreme learning machine based reinforcement learning for goal localization

Nouar AlDahoul, Zaw Zaw Htike, Rini Akmeliawati
2017 IOP Conference Series: Materials Science and Engineering  
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.  ...  Another system that uses raw visual input data is pole balancing controller, it was proposed to learn a control policy by using reinforcement learning [5] .  ... 
doi:10.1088/1757-899x/184/1/012055 fatcat:eqchyfaskrbsbo6dmtz253fl3u

Reinforcement Learning with Videos: Combining Offline Observations with Interaction [article]

Karl Schmeckpeper, Oleh Rybkin, Kostas Daniilidis, Sergey Levine, Chelsea Finn
2021 arXiv   pre-print
In this paper, we consider the question: can we perform reinforcement learning directly on experience collected by humans?  ...  RLV learns a policy and value function using experience collected by humans in combination with data collected by robots.  ...  This work was supported by ARL RCTA W911NF-10-2-0016, ARL DCIST CRA W911NF-17-2-0181, ONR grant N00014-20-1-2675, and by Honda Research Institute.  ... 
arXiv:2011.06507v2 fatcat:twqtxfonrjcpteab4i4z6wmwqq

Page 321 of Psychological Review Vol. 67, Issue 5 [page]

1960 Psychological Review  
(1959) have demonstrated it in discrimination learning by mon- kevs.  ...  For S observes M trying to solve a problem by certain techniques and is more likely to use the same techniques when faced by a similar problem than if M had failed to solve the problem.  ... 

Does observation influence learning?

Olivier Armantier
2004 Games and Economic Behavior  
, and it dominates competing models such as the reinforcement of best response strategies.  2003 Published by Elsevier Science (USA).  ...  A general reinforcement model for continuous strategies, encompassing choice reinforcement learning, direction learning and payoff dependent imitation, performs well in explaining the experimental data  ...  Consider a general reinforcement learning model consisting of L different observational and experiential reinforcement rules.  ... 
doi:10.1016/s0899-8256(03)00124-6 fatcat:ltmlqifnlje4vfdq73jwa3oza4
« Previous Showing results 1 — 15 out of 665,392 results