Return to search

All learning is local: Multi-agent learning in global reward games

In large multiagent games, partial observability, coordination, and credit assignment persistently plague attempts to design good learning algorithms. We provide a simple and efficient algorithm that in part uses a linear system to model the world from a single agent’s limited perspective, and takes advantage of Kalman filtering to allow an agent to construct a good training signal and effectively learn a near-optimal policy in a wide variety of settings. A sequence of increasingly complex empirical tests verifies the efficacy of this technique. / Singapore-MIT Alliance (SMA)

Identiferoai:union.ndltd.org:MIT/oai:dspace.mit.edu:1721.1/3851
Date01 1900
CreatorsChang, Yu-Han, Ho, Tracey, Kaelbling, Leslie P.
Source SetsM.I.T. Theses and Dissertation
Languageen_US
Detected LanguageEnglish
TypeArticle
Format1408858 bytes, application/pdf
RelationComputer Science (CS);

Page generated in 0.0016 seconds