{"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/learning-to-gather-without-communication","title":"Learning to Gather without Communication","arxiv_id":"1802.07834","date":"2018-02-21","proceeding":null,"authors":["El Mahdi El Mhamdi","Rachid Guerraoui","Alexandre Maurer","Vladislav Tempez"],"abstract":"A standard belief on emerging collective behavior is that it emerges from\nsimple individual rules. Most of the mathematical research on such collective\nbehavior starts from imperative individual rules, like always go to the center.\nBut how could an (optimal) individual rule emerge during a short period within\nthe group lifetime, especially if communication is not available. We argue that\nsuch rules can actually emerge in a group in a short span of time via\ncollective (multi-agent) reinforcement learning, i.e learning via rewards and\npunishments. We consider the gathering problem: several agents (social animals,\nswarming robots...) must gather around a same position, which is not determined\nin advance. They must do so without communication on their planned decision,\njust by looking at the position of other agents. We present the first\nexperimental evidence that a gathering behavior can be learned without\ncommunication in a partially observable environment. The learned behavior has\nthe same properties as a self-stabilizing distributed algorithm, as processes\ncan gather from any initial state (and thus tolerate any transient failure).\nBesides, we show that it is possible to tolerate the brutal loss of up to 90\\%\nof agents without significant impact on the behavior.","url_abs":"http://arxiv.org/abs/1802.07834v1","url_pdf":"http://arxiv.org/pdf/1802.07834v1.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":"learning-to-gather-without-communication","repo_url":"https://github.com/LearningToGatherWithoutCommunication/Ring","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"multi-agent-reinforcement-learning","task_name":"Multi-agent Reinforcement Learning"},{"task_slug":null,"task_name":"Position"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}