Improving Reinforcement Learning with Context Detection

Bruno Castro Da Silva (1) , Eduardo W. Basso (1) , Filipo Studzinski Perotto (2) , Ana Lúcia Cetertich Bazzan (1) , Paulo Martins Engel (1)

1 UFRGS - Instituto de Informática da UFRGS
2 IRIT - Institut de recherche en informatique de Toulouse

In this paper we propose a method for solving reinforcement learning problems in non-stationary environments. The basic idea is to create and simultaneously update multiple partial models of the environment dynamics. The learning mechanism is based on the detection of context changes, that is, on the detection of significant changes in the dynamics of the environment. Based on this motivation, we propose, formalize and show the efficiency of a method for detecting the current context and the associated model of prediction, as well as a method for updating each of the incrementally built models.

Mots clés: multimodel learning; non-stationary environments; reinforcement learning

Domaines: Intelligence artificielle [cs.AI]

Bruno Castro Da Silva, Eduardo W. Basso, Filipo Studzinski Perotto, Ana Lúcia Cetertich Bazzan, Paulo Martins Engel. Improving Reinforcement Learning with Context Detection. 5th International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS 2006), Hakodate, Japan, May 8-12, 2006, 2006, New York, NY, USA, United States. pp.810--812, ⟨10.1145/1160633.1160779⟩. ⟨hal-01762261⟩