Kalman Temporal Differences - CentraleSupélec Access content directly
Journal Articles Journal of Artificial Intelligence Research Year : 2010

Kalman Temporal Differences

Abstract

Because reinforcement learning suffers from a lack of scalability, online value (and Q-) function approximation has received increasing interest this last decade. This contribution introduces a novel approximation scheme, namely the Kalman Temporal Differences (KTD) framework, that exhibits the following features: sample-efficiency, non-linear approximation, non-stationarity handling and uncertainty management. A first KTD-based algorithm is provided for deterministic Markov Decision Processes (MDP) which produces biased estimates in the case of stochastic transitions. Than the eXtended KTD framework (XKTD), solving stochastic MDP, is described. Convergence is analyzed for special cases for both deterministic and stochastic transitions. Related algorithms are experimented on classical benchmarks. They compare favorably to the state of the art while exhibiting the announced features.
No file

Dates and versions

hal-00858687 , version 1 (05-09-2013)

Identifiers

  • HAL Id : hal-00858687 , version 1

Cite

Matthieu Geist, Olivier Pietquin. Kalman Temporal Differences. Journal of Artificial Intelligence Research, 2010, 39, pp.483-532. ⟨hal-00858687⟩
59 View
0 Download

Share

Gmail Facebook X LinkedIn More