Error reducing sampling in reinforcement learning

Bruno Scherrer 1 Shie Mannor 2
1 MAIA - Autonomous intelligent machine
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : In reinforcement learning, an agent collects information interacting with an environment and uses it to derive a behavior. This paper focuses on efficient sampling; that is, the problem of choosing the interaction samples so that the corresponding behavior tends quickly to the optimal behavior. Our main result is a sensitivity analysis relating the choice of sampling any state-action pair to the decrease of an error bound on the optimal solution. We derive two new model-based algorithms. Simulations demonstrate a quicker convergence (in the sense of the number of samples) of the value function to the real optimal value function.
Document type :
Reports
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.inria.fr/inria-00098352
Contributor : Bruno Scherrer <>
Submitted on : Monday, September 25, 2006 - 4:13:18 PM
Last modification on : Thursday, January 11, 2018 - 6:19:51 AM
Long-term archiving on : Tuesday, April 6, 2010 - 1:10:20 AM

Identifiers

  • HAL Id : inria-00098352, version 1

Collections

Citation

Bruno Scherrer, Shie Mannor. Error reducing sampling in reinforcement learning. [Research Report] 2004, pp.15. ⟨inria-00098352⟩

Share

Metrics

Record views

197

Files downloads

116