{"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/m3rl-mind-aware-multi-agent-management-1","title":"M$^3$RL: Mind-aware Multi-agent Management Reinforcement Learning","arxiv_id":"1810.00147","date":"2018-09-29","proceeding":"ICLR 2019 5","authors":["Tianmin Shu","Yuandong Tian"],"abstract":"Most of the prior work on multi-agent reinforcement learning (MARL) achieves\noptimal collaboration by directly controlling the agents to maximize a common\nreward. In this paper, we aim to address this from a different angle. In\nparticular, we consider scenarios where there are self-interested agents (i.e.,\nworker agents) which have their own minds (preferences, intentions, skills,\netc.) and can not be dictated to perform tasks they do not wish to do. For\nachieving optimal coordination among these agents, we train a super agent\n(i.e., the manager) to manage them by first inferring their minds based on both\ncurrent and past observations and then initiating contracts to assign suitable\ntasks to workers and promise to reward them with corresponding bonuses so that\nthey will agree to work together. The objective of the manager is maximizing\nthe overall productivity as well as minimizing payments made to the workers for\nad-hoc worker teaming. To train the manager, we propose Mind-aware Multi-agent\nManagement Reinforcement Learning (M^3RL), which consists of agent modeling and\npolicy learning. We have evaluated our approach in two environments, Resource\nCollection and Crafting, to simulate multi-agent management problems with\nvarious task settings and multiple designs for the worker agents. The\nexperimental results have validated the effectiveness of our approach in\nmodeling worker agents' minds online, and in achieving optimal ad-hoc teaming\nwith good generalization and fast adaptation.","url_abs":"http://arxiv.org/abs/1810.00147v3","url_pdf":"http://arxiv.org/pdf/1810.00147v3.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":"m3rl-mind-aware-multi-agent-management-1","repo_url":"https://github.com/facebookresearch/M3RL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"management","task_name":"Management"},{"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":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1810.00147","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}