Filters








16,856 Hits in 5.6 sec

Argumentation Accelerated Reinforcement Learning for RoboCup Keepaway-Takeaway [chapter]

Yang Gao, Francesca Toni
2014 Lecture Notes in Computer Science  
Reinforcement Learning (RL) is widely regarded as a generic and effective technique to learn coordinated behaviours in cooperative multi-agent systems (CMAS), but it suffers from slow convergence speed  ...  We consider a widely used CMAS application, RoboCup Takeaway, and use value-based argumentation to extract heuristics from conflicting domain knowledge therein.  ...  behaviours in cooperative multi-agent systems (CMAS) [1, 2] .  ... 
doi:10.1007/978-3-642-54373-9_6 fatcat:3v3xzthuffeolotlo6lfuvuizq

State Elimination in Accelerated Multiagent Reinforcement Learning

Ary Setijadi Prihatmanto, Widyawardana Adiprawita, Safreni Candra Sari, Kuspriyanto Kuspriyanto
2016 International Journal on Electrical Engineering and Informatics  
This paper presents a novel algorithm of Multiagent Reinforcement Learning called State Elimination in Accelerated Multiagent Reinforcement Learning (SEA-MRL), that successfully produces faster learning  ...  This algorithm is generally applicable for other multiagent task challenges or general multiagent learning with large scale state space, and perfectly applicable with no adjustments for single agent learning  ...  [21] presented a novel class of algorithms, called Heuristically-Accelerated Multi-agent Reinforcement Learning (HAMRL), which allows the use of heuristics to speed up well-known multi-agent reinforcement  ... 
doi:10.15676/ijeei.2016.8.3.12 fatcat:aydhzofht5hfhlypek4xkterpq

A multi-agent cooperative reinforcement learning model using a hierarchy of consultants, tutors and workers

Bilal H. Abed-alguni, Stephan K. Chalup, Frans A. Henskens, David J. Paul
2015 Vietnam Journal of Computer Science  
This paper proposes an algorithm for cooperative policy construction for independent learners, named Q-learning with aggregation (QA-learning).  ...  The hierarchical organisation of distributed systems can provide an efficient decomposition for machine learning.  ...  [8] proposed a combined hierarchical reinforcement learning method for multi-robot cooperation in completely unknown environments.  ... 
doi:10.1007/s40595-015-0045-x fatcat:aucrigrh3fherni67zpio7zkue

Cooperative Multi-Agent Learning: The State of the Art

Liviu Panait, Sean Luke
2005 Autonomous Agents and Multi-Agent Systems  
Cooperative multi-agent systems are ones in which several agents attempt, through their interaction, to jointly solve tasks or to maximize utility.  ...  We provide a broad survey of the cooperative multi-agent learning literature.  ...  Except for section 3.2.3, the survey primarily covers topics of interest to cooperative multi-agent learning.  ... 
doi:10.1007/s10458-005-2631-2 fatcat:u3xlftotajfitdtfmvbmggwgbi

Combining Dynamic Reward Shaping and Action Shaping for Coordinating Multi-agent Learning

Xiangbin Zhu, Chongjie Zhang, Victor Lesser
2013 2013 IEEE/WIC/ACM International Joint Conferences on Web Intelligence (WI) and Intelligent Agent Technologies (IAT)  
Coordinating multi-agent reinforcement learning provides a promising approach to scaling learning in large cooperative multi-agent systems.  ...  Experimental results show our approach effectively speeds up the convergence of multi-agent learning in large systems.  ...  INTRODUCTION A central question in developing cooperative multi-agent systems is to design distributed coordination policies for agents so that they work together to optimize the global system performance  ... 
doi:10.1109/wi-iat.2013.127 dblp:conf/iat/ZhuZL13 fatcat:3cqy76gnjvapjdkk5chuonydma

Multi-Agent Reinforcement Learning: A Survey

Lucian Busoniu, Robert Babuska, Bart De Schutter
2006 2006 9th International Conference on Control, Automation, Robotics and Vision  
Many tasks arising in these domains require that the agents learn behaviors online. A significant part of the research on multi-agent learning concerns reinforcement learning techniques.  ...  Multi-agent systems are rapidly finding applications in a variety of domains, including robotics, distributed control, telecommunications, economics.  ...  ACKNOWLEDGEMENT This research is financially supported by Senter, Ministry of Economic Affairs of the Netherlands within the BSIK-ICIS project "Interactive Collaborative Information Systems" (grant no.  ... 
doi:10.1109/icarcv.2006.345353 dblp:conf/icarcv/BusoniuBS06 fatcat:5lo6wzdlbncybbb2uu4bdrzsqq

Evolutionary Game Theory Based Cooperation Algorithm in Multi-Agent System [chapter]

Yuehai Wang
2009 Multiagent Systems  
www.intechopen.com Evolutionary Game Theory based Cooperation Algorithm in Multi-agent System www.intechopen.com  ...  Game theory based cooperation approach for multi-agent system 4.1 The relationship between the optimal cooperation solution of MAS and Nash equilibrium of the corresponding game To accomplish a mission  ...  Evolutionary Game Theory Based Cooperation Algorithm in Multi-Agent System, Multiagent Systems, Salman Ahmed and Mohd Noh Karsiti (Ed.), ISBN: 978-3-902613-51-6, InTech, Available from: http://www.intechopen.com  ... 
doi:10.5772/6601 fatcat:khjs7w3npvczzmgg3u3qxgzso4

CM3: Cooperative Multi-goal Multi-stage Multi-agent Reinforcement Learning [article]

Jiachen Yang, Alireza Nakhaei, David Isele, Kikuo Fujimura, Hongyuan Zha
2020 arXiv   pre-print
This multi-goal multi-agent setting poses difficulties for recent algorithms, which primarily target settings with a single global reward, due to two new challenges: efficient exploration for learning  ...  To address both challenges, we restructure the problem into a novel two-stage curriculum, in which single-agent goal attainment is learned prior to learning multi-agent cooperation, and we derive a new  ...  ACKNOWLEDGMENTS JY thanks Rakshit Trivedi (Georgia Institute of Technology), Ahmad Beirami (Facebook AI), and Peter Sunehag (DeepMind) for detailed and helpful feedback on this work.  ... 
arXiv:1809.05188v3 fatcat:dld5q6ycojcx7elo5ugou26bti

Reward Design for Multi-Agent Reinforcement Learning with a Penalty Based on the Payment Mechanism

Natsuki Matsunami, Shun Okuhara, Takayuki Ito
2021 Transactions of the Japanese society for artificial intelligence  
In this paper, we propose a novel method of reward design for multi-agent reinforcement learning (MARL). One of the main uses of MARL is building cooperative policies between self-interested agents.  ...  We give the individual learning agent a reward signal that consists of two elements.  ...  Rafik Hadfi for comments that greatly improved the manuscript.  ... 
doi:10.1527/tjsai.36-5_ag21-h fatcat:xeo7gzunfvh7nkxxil7kstlznm

Decentralized Reinforcement Learning of Robot Behaviors

David L. Leottau, Javier Ruiz-del-Solar, Robert Babuška
2018 Artificial Intelligence  
A multi-agent methodology is proposed for Decentralized Reinforcement Learning (DRL) of individual behaviors in problems where multi-dimensional action spaces are involved.  ...  In this paper, three Multi-Agent Learning (MAL) algorithms are considered and tested: the independent DRL, the Cooperative Adaptive (CA) Learning Rate, and a Lenient learning approach extended to multi-state  ...  (f) Lenient Multi-Agent Reinforcement Learning [35] : it showed asymptotic convergence when applied to multi-state DRL problems.  ... 
doi:10.1016/j.artint.2017.12.001 fatcat:szutkob4kfbypdsc4gfogwb2a4

Scaling Up Multiagent Reinforcement Learning for Robotic Systems: Learn an Adaptive Sparse Communication Graph [article]

Chuangchuang Sun, Macheng Shen, Jonathan P. How
2020 arXiv   pre-print
The complexity of multiagent reinforcement learning (MARL) in multiagent systems increases exponentially with respect to the agent number.  ...  Comparative results show that our algorithm can learn an interpretable sparse structure and outperforms previous works by a significant margin on applications involving a large-scale multiagent system.  ...  ACKNOWLEDGMENTS Research is supported by Scientific Systems Company, Inc. under research agreement # SC-1661-04.  ... 
arXiv:2003.01040v2 fatcat:ch3zgjpd2vcmbjpfis5nm3gglm

The Neural MMO Platform for Massively Multiagent Research [article]

Joseph Suarez, Yilun Du, Clare Zhu, Igor Mordatch, Phillip Isola
2021 arXiv   pre-print
Initial baselines on the platform demonstrate that agents trained in large populations explore more and learn a progression of skills.  ...  Neural MMO is a computationally accessible research platform that combines large agent populations, long time horizons, open-ended tasks, and modular game systems.  ...  It is less suitable for benchmarking reinforcement learning algorithms because agent performance against clones is not indicative of overall policy quality.  ... 
arXiv:2110.07594v1 fatcat:ev7skquzwvhuxeux46cr7eaej4

Dynamic spectrum access and sharing through actor-critic deep reinforcement learning

Liang Dong, Yuchen Qian, Yuan Xing
2022 EURASIP Journal on Wireless Communications and Networking  
In addition, multiple secondary users implement multi-agent deep reinforcement learning under certain coordination.  ...  learning.  ...  When new agents are inserted into the system, they can quickly join the cooperation of multi-agent reinforcement learning. Therefore, the algorithm is inherently robust.  ... 
doi:10.1186/s13638-022-02124-4 fatcat:2fz7hdbrefgohhf6lcim67wrkm

Cooperative Autonomous Vehicles that Sympathize with Human Drivers [article]

Behrad Toghi, Rodolfo Valiente, Dorsa Sadigh, Ramtin Pedarsani, Yaser P. Fallah
2021 arXiv   pre-print
In contrast with existing works that explicitly model the behavior of human drivers and rely on their expected response to create opportunities for cooperation, our Sympathetic Cooperative Driving (SymCoDrive  ...  ) paradigm trains altruistic agents that realize safe and smooth traffic flow in competitive driving scenarios only from experiential learning and without any explicit coordination.  ...  RELATED WORK Multi-agent Reinforcement Learning. A major challenge for multi-agent reinforcement learning (MARL) is the in-herent non-stationarity of the environment.  ... 
arXiv:2107.00898v1 fatcat:diskrjyt25bhzfq3hcvkdguehe

Social Coordination and Altruism in Autonomous Driving [article]

Behrad Toghi, Rodolfo Valiente, Dorsa Sadigh, Ramtin Pedarsani, Yaser P. Fallah
2022 arXiv   pre-print
reinforcement learning framework.  ...  Formally, we model an AV's maneuver planning in mixed-autonomy traffic as a partially-observable stochastic game and attempt to derive optimal policies that lead to socially-desirable outcomes using a multi-agent  ...  Multi-agent Reinforcement Learning. Early solutions for multi-agent value-learning algorithms assume independently trained agents and are proved to perform poorly [7] .  ... 
arXiv:2107.00200v4 fatcat:tefw5v7dqvfh7ebdjghnaakzh4
« Previous Showing results 1 — 15 out of 16,856 results