963 Hits in 3.6 sec

Partial Local FriendQ Multiagent Learning: Application to Team Automobile Coordination Problem [chapter]

Julien Laumonier, Brahim Chaib-draa
2006 Lecture Notes in Computer Science  
Real world multiagent coordination problems are important issues for reinforcement learning techniques.  ...  If each degree of observability is associated with communication costs, multiagent system designers are able to choose a compromise between the performance of the policy and the cost to obtain the associated  ...  plan to discover others kind of distance between observability to generalize our approach to positive and negative interaction management problems in teams.  ... 
doi:10.1007/11766247_31 fatcat:tmentiqzerfdlcj5yhbyg2gdr4

Hierarchical Deep Multiagent Reinforcement Learning with Temporal Abstraction [article]

Hongyao Tang, Jianye Hao, Tangjie Lv, Yingfeng Chen, Zongzhang Zhang, Hangtian Jia, Chunxu Ren, Yan Zheng, Zhaopeng Meng, Changjie Fan, Li Wang
2019 arXiv   pre-print
With temporal abstraction, we decompose the problem into a hierarchy of different time scales and investigate how agents can learn high-level coordination based on the independent skills learned at the  ...  In this paper, we study hierarchical deep MARL in cooperative multiagent problems with sparse and delayed reward.  ...  With Deep Q-Learning (Mnih et al. 2015), a high-level policy π(g t |s t ) can be learned by minimizing the loss with parameters θ, L(θ) = E st,gt,τ,rt,..t+τ ,st+τ y − Q θ (s t , g t ) 2 , (1) where y =  ... 
arXiv:1809.09332v2 fatcat:ii73dmrohvhihklyga6xun2qgq

Modeling multidisciplinary design with multiagent learning

Daniel Hulse, Kagan Tumer, Christopher Hoyle, Irem Tumer
2018 Artificial intelligence for engineering design, analysis and manufacturing  
This situation is analogous to a multiagent system in which agents solve individual parts of a larger problem in a coordinated way.  ...  These designers are represented as a multiagent learning system which is shown to perform similarly to a centralized optimization algorithm on the same domain.  ...  The Q-learning table assignment is: Q(s t , a t ) Q(s t , a t ) + a × (r t + g × max(Q(s t+1 , a t+1 )) − Q(s t , a t )) where Q(s t , a t ) is the value given to the action in the current state, α is  ... 
doi:10.1017/s0890060418000161 fatcat:rp4brvl3mje25pr4s4gnl2yj4u

Evolutionary Reinforcement Learning for Sample-Efficient Multiagent Coordination [article]

Shauharda Khadka and Somdeb Majumdar and Santiago Miret and Stephen McAleer and Kagan Tumer
2020 arXiv   pre-print
Many cooperative multiagent reinforcement learning environments provide agents with a sparse team-based reward, as well as a dense agent-specific reward that incentivizes learning basic skills.  ...  Furthermore, relying solely on the agent-specific reward is sub-optimal because it usually does not capture the team coordination objective.  ...  learn team formation and effective coordination.  ... 
arXiv:1906.07315v3 fatcat:45qh5okclffe5cz7sc2bjhfajm

Learning Hierarchical Teaching Policies for Cooperative Agents [article]

Dong-Ki Kim, Miao Liu, Shayegan Omidshafiei, Sebastian Lopez-Cot, Matthew Riemer, Golnaz Habibi, Gerald Tesauro, Sami Mourad, Murray Campbell, Jonathan P. How
2020 arXiv   pre-print
In particular, recent work studying agents that learn to teach other teammates has demonstrated that action advising accelerates team-wide learning.  ...  Collective learning can be greatly enhanced when agents effectively exchange knowledge with their peers.  ...  Recent work applied imitation learning for multiagent coordination [19] and explored effective combinations of imitation learning and HRL [18] .  ... 
arXiv:1903.03216v6 fatcat:4qz2i5c2k5a3fklmfzxz4r4qoa

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.  ...  Moreover, in a sub-team, the algorithm learns a communication graph similar to a star-graph. It can be understood that each sub-team selects a leader.  ... 
arXiv:2003.01040v2 fatcat:ch3zgjpd2vcmbjpfis5nm3gglm

Action Semantics Network: Considering the Effects of Actions in Multiagent Systems [article]

Weixun Wang, Tianpei Yang, Yong Liu, Jianye Hao, Xiaotian Hao, Yujing Hu, Yingfeng Chen, Changjie Fan, Yang Gao
2020 arXiv   pre-print
Previous works borrow various multiagent coordination mechanisms into deep learning architecture to facilitate multiagent coordination.  ...  ASN can be easily combined with existing deep reinforcement learning (DRL) algorithms to boost their performance.  ...  One major class of works incorporates various multiagent coordination mechanisms into deep multiagent learning architecture (Lowe et al., 2017; Yang et al., 2018; Palmer et al., 2018) .  ... 
arXiv:1907.11461v3 fatcat:3ovjg7ebcjhqrhvcl4a5jkfljq

Independent Generative Adversarial Self-Imitation Learning in Cooperative Multiagent Systems [article]

Xiaotian Hao, Weixun Wang, Jianye Hao, Yaodong Yang
2019 arXiv   pre-print
To the best of our knowledge, we are the first to combine self-imitation learning with generative adversarial imitation learning (GAIL) and apply it to cooperative multiagent systems.  ...  In this paper, we propose a novel framework called Independent Generative Adversarial Self-Imitation Learning (IGASIL) to address the coordination problems in fully cooperative multiagent environments.  ...  However, directly applying single-agent reinforcement learning approaches such as Q-learning to cooperative multiagent environments behaves poorly.  ... 
arXiv:1909.11468v1 fatcat:glwziahczfcnbfkzgw4itviuli

Scalable multiagent learning through indirect encoding of policy geometry

David B. D'Ambrosio, Kenneth O. Stanley
2013 Evolutionary Intelligence  
This paper presents an alternative approach to multiagent learning called multiagent HyperNEAT that represents the team as a pattern of policies rather than as a set of individual agents.  ...  In this paper, multiagent HyperNEAT is compared to a traditional learning method, multiagent Sarsa(λ ), in a predator-prey domain, where it demonstrates its ability to train large teams.  ...  Layered learning [45, 98] takes inspiration from multiagent learning in the sense that a complex problem is broken into smaller sub-problems that should be easier to learn.  ... 
doi:10.1007/s12065-012-0086-3 fatcat:7ropvv3r5nemlh4hkliqeymqgi

Dynamic Partition of Collaborative Multiagent Based on Coordination Trees [chapter]

Fang Min, Frans C. A. Groen, Li Hao
2013 Advances in Intelligent Systems and Computing  
The results of the experiments prove that this method outperforms related multi-agent reinforcement-learning methods based on alterable collaborative teams.  ...  We present a coordination tree structure whose nodes are agent subsets or an agent. Two kinds of weights of a tree are defined which describe the cost of an agent collaborating with an agent subset.  ...  The attenuation factor is set to 0. 6 fig.3 , the Q-learning of team Markov games based on coordination trees could converge to the optimal value finally.  ... 
doi:10.1007/978-3-642-33932-5_46 fatcat:obbceljjszf67acrvzjuzfchn4

Exploring the Benefits of Teams in Multiagent Learning [article]

David Radke, Kate Larson, Tim Brecht
2022 arXiv   pre-print
Multiagent teams are primarily studied when in conflict; however, organizational psychology (OP) highlights the benefits of teams among human populations for learning how to coordinate and cooperate.  ...  In this paper, we propose a new model of multiagent teams for reinforcement learning (RL) agents inspired by OP and early work on teams in artificial intelligence.  ...  IPD Evaluation In the IPD, each experiment lasts 1.0 × 10 6 episodes where N = 30 agents learn using Deep Q-Learning [Mnih et al., 2015 ].  ... 
arXiv:2205.02328v1 fatcat:wferjbmrm5a75pmwmgrg3qw6dq

Can bounded and self-interested agents be teammates? Application to planning in ad hoc teams

Muthukumaran Chandrasekaran, Prashant Doshi, Yifeng Zeng, Yingke Chen
2016 Autonomous Agents and Multi-Agent Systems  
Planning for ad hoc teamwork is challenging because it involves agents collaborating without any prior coordination or communication.  ...  We address this limitation by including models at level 0 whose solutions involve reinforcement learning. We show how the learning is integrated into planning in the context of I-DIDs.  ...  Recently, interest in spontaneously coordinating with others as part of an ad hoc (or impromptu) multiagent team is building, driven by applications such as robot soccer [21] and disaster response.  ... 
doi:10.1007/s10458-016-9354-4 fatcat:5vca5wsoenhsteassqszkla5oa

Emergence of multiagent spatial coordination strategies through artificial coevolution

André L.V. Coelho, Daniel Weingaertner, Ricardo R. Gudwin, Ivan L.M. Ricarte
2001 Computers & graphics  
This paper describes research investigating the evolution of coordination strategies in robot soccer teams.  ...  The main results show that, through coevolution, we progressively create teams whose members act on complementary areas of the playing field, being capable of prevailing over a standard opponent team with  ...  In his work, all agents have a common set of skills from which they build their tasks achieving strategy through a Q-learning (reinforcement learning) algorithm.  ... 
doi:10.1016/s0097-8493(01)00155-8 fatcat:4ibwnulwqrepnbvwfp43ij5fue

Adaptive learning: A new decentralized reinforcement learning approach for cooperative multiagent systems

Li Meng-Lin, Chen Shao-Fei, Chen Jing
2020 IEEE Access  
The new algorithm can dynamically adjust the learning rate by deeply analyzing the dissonance problem in the matrix game and combining it with a multiagent environment.  ...  Moreover, the variance of the training results is more stable than that of the hysteretic Q learning(HQL) algorithm.  ...  Thus, it is demonstrated that the adaptive Q-learning algorithm successfully achieves the goal of coordination in various IL-based multiagent scenarios.  ... 
doi:10.1109/access.2020.2997899 fatcat:nyhq2q6u5jhrpknp634xro3wja

Coordination in multiagent reinforcement learning systems by virtual reinforcement signals

M.A.S. Kamal, Junichi Murata
2007 Journal of Knowledge-based & Intelligent Engineering Systems  
This paper presents a novel method for on-line coordination in multiagent reinforcement learning systems.  ...  The empirical performance of the coordinated system compared to the uncoordinated system illustrates its advantages for multiagent systems.  ...  Q(s t , a t ) ← (1 − α)Q(s t , a t ) +α[r t+1 + γ max a Q(s t+1 , a)], Coordinated multiagent reinforcement learning Agents in an MAS can follow the above RL rules independently and concentrate on their  ... 
doi:10.3233/kes-2007-11304 fatcat:fn2nqzzzrrfbnahicnz6tbexny
« Previous Showing results 1 — 15 out of 963 results