{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/fully-decentralized-multi-agent-reinforcement","title":"Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents","arxiv_id":"1802.08757","date":"2018-02-23","proceeding":"ICML 2018 7","authors":["Kaiqing Zhang","Zhuoran Yang","Han Liu","Tong Zhang","Tamer Başar"],"abstract":"We consider the problem of \\emph{fully decentralized} multi-agent\nreinforcement learning (MARL), where the agents are located at the nodes of a\ntime-varying communication network. Specifically, we assume that the reward\nfunctions of the agents might correspond to different tasks, and are only known\nto the corresponding agent. Moreover, each agent makes individual decisions\nbased on both the information observed locally and the messages received from\nits neighbors over the network. Within this setting, the collective goal of the\nagents is to maximize the globally averaged return over the network through\nexchanging information with their neighbors. To this end, we propose two\ndecentralized actor-critic algorithms with function approximation, which are\napplicable to large-scale MARL problems where both the number of states and the\nnumber of agents are massively large. Under the decentralized structure, the\nactor step is performed individually by each agent with no need to infer the\npolicies of others. For the critic step, we propose a consensus update via\ncommunication over the network. Our algorithms are fully incremental and can be\nimplemented in an online fashion. Convergence analyses of the algorithms are\nprovided when the value functions are approximated within the class of linear\nfunctions. Extensive simulation results with both linear and nonlinear function\napproximations are presented to validate the proposed algorithms. Our work\nappears to be the first study of fully decentralized MARL algorithms for\nnetworked agents with function approximation, with provable convergence\nguarantees.","url_abs":"http://arxiv.org/abs/1802.08757v2","url_pdf":"http://arxiv.org/pdf/1802.08757v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"fully-decentralized-multi-agent-reinforcement","repo_url":"https://github.com/MUmarJaved/MultiAgent-Distributed-Reinforcement-Learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"fully-decentralized-multi-agent-reinforcement","repo_url":"https://github.com/anonymous1234517/code","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"fully-decentralized-multi-agent-reinforcement","repo_url":"https://github.com/cts198859/deeprl_dist","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"fully-decentralized-multi-agent-reinforcement","repo_url":"https://github.com/cts198859/deeprl_network","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"fully-decentralized-multi-agent-reinforcement","repo_url":"https://github.com/dongchen06/macacc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"multi-agent-reinforcement-learning","task_name":"Multi-agent Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.08757","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}