{"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-framework-for-factorizing","title":"A Unified Framework for Factorizing Distributional Value Functions for Multi-Agent Reinforcement Learning","arxiv_id":"2306.02430","date":"2023-06-04","proceeding":null,"authors":["Wei-Fang Sun","Cheng-Kuang Lee","Simon See","Chun-Yi Lee"],"abstract":"In fully cooperative multi-agent reinforcement learning (MARL) settings, environments are highly stochastic due to the partial observability of each agent and the continuously changing policies of other agents. To address the above issues, we proposed a unified framework, called DFAC, for integrating distributional RL with value function factorization methods. This framework generalizes expected value function factorization methods to enable the factorization of return distributions. To validate DFAC, we first demonstrate its ability to factorize the value functions of a simple matrix game with stochastic rewards. Then, we perform experiments on all Super Hard maps of the StarCraft Multi-Agent Challenge and six self-designed Ultra Hard maps, showing that DFAC is able to outperform a number of baselines.","url_abs":"https://arxiv.org/abs/2306.02430v1","url_pdf":"https://arxiv.org/pdf/2306.02430v1.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-framework-for-factorizing","repo_url":"https://github.com/j3soon/dfac-extended","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"multi-agent-reinforcement-learning","task_name":"Multi-agent Reinforcement Learning"},{"task_slug":"smac","task_name":"SMAC"},{"task_slug":"starcraft","task_name":"Starcraft"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/smac-on-smac-26m-vs-30m","task":"SMAC","dataset":"SMAC 26m_vs_30m","model":"DMIX","rank_in_archive_order":1,"of":6,"metrics":{"Average Score":"19.17","Median Win Rate":"81.82"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-26m-vs-30m","task":"SMAC","dataset":"SMAC 26m_vs_30m","model":"QPLEX","rank_in_archive_order":2,"of":6,"metrics":{"Average Score":"18.66","Median Win Rate":"78.12"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-26m-vs-30m","task":"SMAC","dataset":"SMAC 26m_vs_30m","model":"DDN","rank_in_archive_order":3,"of":6,"metrics":{"Average Score":"18.49","Median Win Rate":"67.90"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-26m-vs-30m","task":"SMAC","dataset":"SMAC 26m_vs_30m","model":"DPLEX","rank_in_archive_order":4,"of":6,"metrics":{"Average Score":"18.49","Median Win Rate":"59.38"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-26m-vs-30m","task":"SMAC","dataset":"SMAC 26m_vs_30m","model":"QMIX","rank_in_archive_order":5,"of":6,"metrics":{"Average Score":"18.23","Median Win Rate":"62.78"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-26m-vs-30m","task":"SMAC","dataset":"SMAC 26m_vs_30m","model":"VDN","rank_in_archive_order":6,"of":6,"metrics":{"Average Score":"16.69","Median Win Rate":"23.01"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-27m-vs-30m","task":"SMAC","dataset":"SMAC 27m_vs_30m","model":"DPLEX","rank_in_archive_order":2,"of":11,"metrics":{"Average Score":"19.62","Median Win Rate":"90.62"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-27m-vs-30m","task":"SMAC","dataset":"SMAC 27m_vs_30m","model":"QPLEX","rank_in_archive_order":5,"of":11,"metrics":{"Average Score":"19.33","Median Win Rate":"78.12"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-3s6z-1","task":"SMAC","dataset":"SMAC 3s5z_vs_3s6z","model":"DPLEX","rank_in_archive_order":4,"of":13,"metrics":{"Average Score":"20.27","Median Win Rate":"90.62"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-3s6z-1","task":"SMAC","dataset":"SMAC 3s5z_vs_3s6z","model":"QPLEX","rank_in_archive_order":6,"of":13,"metrics":{"Average Score":"20.42","Median Win Rate":"84.38"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-4s6z","task":"SMAC","dataset":"SMAC 3s5z_vs_4s6z","model":"DDN","rank_in_archive_order":1,"of":6,"metrics":{"Average Score":"19.65","Median Win Rate":"89.77"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-4s6z","task":"SMAC","dataset":"SMAC 3s5z_vs_4s6z","model":"DMIX","rank_in_archive_order":2,"of":6,"metrics":{"Average Score":"18.61","Median Win Rate":"83.52"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-4s6z","task":"SMAC","dataset":"SMAC 3s5z_vs_4s6z","model":"VDN","rank_in_archive_order":3,"of":6,"metrics":{"Average Score":"17.16","Median Win Rate":"47.16"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-4s6z","task":"SMAC","dataset":"SMAC 3s5z_vs_4s6z","model":"DPLEX","rank_in_archive_order":4,"of":6,"metrics":{"Average Score":"14.99"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-4s6z","task":"SMAC","dataset":"SMAC 3s5z_vs_4s6z","model":"QPLEX","rank_in_archive_order":5,"of":6,"metrics":{"Average Score":"13.60"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-3s5z-vs-4s6z","task":"SMAC","dataset":"SMAC 3s5z_vs_4s6z","model":"QMIX","rank_in_archive_order":6,"of":6,"metrics":{"Average Score":"13.09"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-8z-1","task":"SMAC","dataset":"SMAC 6h_vs_8z","model":"DPLEX","rank_in_archive_order":4,"of":14,"metrics":{"Average Score":"17.88","Median Win Rate":"43.75"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-8z-1","task":"SMAC","dataset":"SMAC 6h_vs_8z","model":"QPLEX","rank_in_archive_order":14,"of":14,"metrics":{"Average Score":"15.95"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-9z","task":"SMAC","dataset":"SMAC 6h_vs_9z","model":"DDN","rank_in_archive_order":1,"of":6,"metrics":{"Average Score":"16.00","Median Win Rate":"0.28"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-9z","task":"SMAC","dataset":"SMAC 6h_vs_9z","model":"DPLEX","rank_in_archive_order":2,"of":6,"metrics":{"Average Score":"14.84"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-9z","task":"SMAC","dataset":"SMAC 6h_vs_9z","model":"QPLEX","rank_in_archive_order":3,"of":6,"metrics":{"Average Score":"13.86"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-9z","task":"SMAC","dataset":"SMAC 6h_vs_9z","model":"DMIX","rank_in_archive_order":4,"of":6,"metrics":{"Average Score":"13.73"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-9z","task":"SMAC","dataset":"SMAC 6h_vs_9z","model":"VDN","rank_in_archive_order":5,"of":6,"metrics":{"Average Score":"13.57"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-6h-vs-9z","task":"SMAC","dataset":"SMAC 6h_vs_9z","model":"QMIX","rank_in_archive_order":6,"of":6,"metrics":{"Average Score":"12.37","Median Win Rate":"1.14"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-1","task":"SMAC","dataset":"SMAC MMM2","model":"DPLEX","rank_in_archive_order":3,"of":14,"metrics":{"Average Score":"19.93","Median Win Rate":"96.88"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-1","task":"SMAC","dataset":"SMAC MMM2","model":"QPLEX","rank_in_archive_order":4,"of":14,"metrics":{"Average Score":"19.60","Median Win Rate":"96.88"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-8m4m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_8m4M1M","model":"DDN","rank_in_archive_order":1,"of":6,"metrics":{"Average Score":"16.50","Median Win Rate":"56.82"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-8m4m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_8m4M1M","model":"DMIX","rank_in_archive_order":2,"of":6,"metrics":{"Average Score":"16.24","Median Win Rate":"63.35"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-8m4m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_8m4M1M","model":"DPLEX","rank_in_archive_order":3,"of":6,"metrics":{"Average Score":"15.89","Median Win Rate":"50.00"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-8m4m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_8m4M1M","model":"QPLEX","rank_in_archive_order":4,"of":6,"metrics":{"Average Score":"15.52","Median Win Rate":"46.88"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-8m4m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_8m4M1M","model":"QMIX","rank_in_archive_order":5,"of":6,"metrics":{"Average Score":"14.40","Median Win Rate":"29.55"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-8m4m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_8m4M1M","model":"VDN","rank_in_archive_order":6,"of":6,"metrics":{"Average Score":"13.13","Median Win Rate":"13.35"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-9m3m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_9m3M1M","model":"DDN","rank_in_archive_order":1,"of":6,"metrics":{"Average Score":"19.45","Median Win Rate":"90.34"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-9m3m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_9m3M1M","model":"DPLEX","rank_in_archive_order":2,"of":6,"metrics":{"Average Score":"19.40","Median Win Rate":"90.62"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-9m3m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_9m3M1M","model":"DMIX","rank_in_archive_order":3,"of":6,"metrics":{"Average Score":"19.33","Median Win Rate":"92.33"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-9m3m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_9m3M1M","model":"QPLEX","rank_in_archive_order":4,"of":6,"metrics":{"Average Score":"19.06","Median Win Rate":"90.62"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-9m3m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_9m3M1M","model":"QMIX","rank_in_archive_order":5,"of":6,"metrics":{"Average Score":"19.01","Median Win Rate":"88.64"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-mmm2-7m2m1m-vs-9m3m1m","task":"SMAC","dataset":"SMAC MMM2_7m2M1M_vs_9m3M1M","model":"VDN","rank_in_archive_order":6,"of":6,"metrics":{"Average Score":"17.30","Median Win Rate":"75.00"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor","task":"SMAC","dataset":"SMAC corridor","model":"DPLEX","rank_in_archive_order":7,"of":13,"metrics":{"Average Score":"19.08","Median Win Rate":"81.25"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor","task":"SMAC","dataset":"SMAC corridor","model":"QPLEX","rank_in_archive_order":8,"of":13,"metrics":{"Average Score":"18.73","Median Win Rate":"75.00"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor-2z-vs-24zg","task":"SMAC","dataset":"SMAC corridor_2z_vs_24zg","model":"DDN","rank_in_archive_order":1,"of":6,"metrics":{"Average Score":"11.10","Median Win Rate":"41.19"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor-2z-vs-24zg","task":"SMAC","dataset":"SMAC corridor_2z_vs_24zg","model":"DPLEX","rank_in_archive_order":2,"of":6,"metrics":{"Average Score":"10.71","Median Win Rate":"3.12"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor-2z-vs-24zg","task":"SMAC","dataset":"SMAC corridor_2z_vs_24zg","model":"VDN","rank_in_archive_order":3,"of":6,"metrics":{"Average Score":"7.78","Median Win Rate":"0.00"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor-2z-vs-24zg","task":"SMAC","dataset":"SMAC corridor_2z_vs_24zg","model":"DMIX","rank_in_archive_order":4,"of":6,"metrics":{"Average Score":"7.41"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor-2z-vs-24zg","task":"SMAC","dataset":"SMAC corridor_2z_vs_24zg","model":"QPLEX","rank_in_archive_order":5,"of":6,"metrics":{"Average Score":"6.44"},"uses_additional_data":false},{"leaderboard":"/sota/smac-on-smac-corridor-2z-vs-24zg","task":"SMAC","dataset":"SMAC corridor_2z_vs_24zg","model":"QMIX","rank_in_archive_order":6,"of":6,"metrics":{"Average Score":"4.80"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}