{"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/a-unified-game-theoretic-approach-to","title":"A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning","arxiv_id":"1711.00832","date":"2017-11-02","proceeding":"NeurIPS 2017 12","authors":["Marc Lanctot","Vinicius Zambaldi","Audrunas Gruslys","Angeliki Lazaridou","Karl Tuyls","Julien Perolat","David Silver","Thore Graepel"],"abstract":"To achieve general intelligence, agents must learn how to interact with\nothers in a shared environment: this is the challenge of multiagent\nreinforcement learning (MARL). The simplest form is independent reinforcement\nlearning (InRL), where each agent treats its experience as part of its\n(non-stationary) environment. In this paper, we first observe that policies\nlearned using InRL can overfit to the other agents' policies during training,\nfailing to sufficiently generalize during execution. We introduce a new metric,\njoint-policy correlation, to quantify this effect. We describe an algorithm for\ngeneral MARL, based on approximate best responses to mixtures of policies\ngenerated using deep reinforcement learning, and empirical game-theoretic\nanalysis to compute meta-strategies for policy selection. The algorithm\ngeneralizes previous ones such as InRL, iterated best response, double oracle,\nand fictitious play. Then, we present a scalable implementation which reduces\nthe memory requirement using decoupled meta-solvers. Finally, we demonstrate\nthe generality of the resulting policies in two partially observable settings:\ngridworld coordination games and poker.","url_abs":"http://arxiv.org/abs/1711.00832v2","url_pdf":"http://arxiv.org/pdf/1711.00832v2.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":"a-unified-game-theoretic-approach-to","repo_url":"https://github.com/deepmind/open_spiel","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep 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=1711.00832","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}