Multi-agent learning

Multi-agent learning is the use of machine learning in a multi-agent system.[1] Typically, agents improve their decisions via experience. In particular, an agent has to learn how to coordinate with the other agents.

Overview

According to an article by Shoham et al. in 2007, it is difficult to pinpoint all relevant articles in the domain.[2] There are some inherent difficulties about multi-agent deep reinforcement learning.[3] The environment is not stationnary anymore, thus the Markov property is violated: transitions and rewards does not only depend on the current state of an agent.

References
  1. Albrecht, Stefano; Stone, Peter (2017), "Multiagent Learning: Foundations and Recent Trends. Tutorial", IJCAI-17 conference (PDF)
  2. Shoham, Yoav; Powers, Rob; Grenager, Trond (2007-05-01). "If multi-agent learning is the answer, what is the question?". Artificial Intelligence. Foundations of Multi-Agent Learning. 171 (7): 365–377. doi:10.1016/j.artint.2006.02.006. ISSN 0004-3702.
  3. Hernandez-Leal, Pablo; Kartal, Bilal; Taylor, Matthew E. (2019-11-01). "A survey and critique of multiagent deep reinforcement learning". Autonomous Agents and Multi-Agent Systems. 33 (6): 750–797. arXiv:1810.05587. doi:10.1007/s10458-019-09421-1. ISSN 1573-7454.


This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.