Anderson acceleration for reinforcement learning

Abstract : Anderson (1965) acceleration is an old and simple method for accelerating the computation of a fixed point. However, as far as we know and quite surprisingly, it has never been applied to dynamic programming or reinforcement learning. In this paper, we explain briefly what Anderson acceleration is and how it can be applied to value iteration, this being supported by preliminary experiments showing a significant speed up of convergence, that we critically discuss. We also discuss how this idea could be applied more generally to (deep) reinforcement learning.
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download

https://hal.inria.fr/hal-01928142
Contributor : Bruno Scherrer <>
Submitted on : Tuesday, November 20, 2018 - 12:52:47 PM
Last modification on : Monday, November 26, 2018 - 2:00:08 PM

File

ewrl_14_2018_paper_1.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01928142, version 1
  • ARXIV : 1809.09501

Collections

Citation

Matthieu Geist, Bruno Scherrer. Anderson acceleration for reinforcement learning. EWRL 2018 - 4th European workshop on Reinforcement Learning, Oct 2018, Lille, France. ⟨hal-01928142⟩

Share

Metrics

Record views

122

Files downloads

77