English

A Tighter Convergence Proof of Reverse Experience Replay

Machine Learning 2024-09-02 v1 Machine Learning

Abstract

In reinforcement learning, Reverse Experience Replay (RER) is a recently proposed algorithm that attains better sample complexity than the classic experience replay method. RER requires the learning algorithm to update the parameters through consecutive state-action-reward tuples in reverse order. However, the most recent theoretical analysis only holds for a minimal learning rate and short consecutive steps, which converge slower than those large learning rate algorithms without RER. In view of this theoretical and empirical gap, we provide a tighter analysis that mitigates the limitation on the learning rate and the length of consecutive steps. Furthermore, we show theoretically that RER converges with a larger learning rate and a longer sequence.

Keywords

Cite

@article{arxiv.2408.16999,
  title  = {A Tighter Convergence Proof of Reverse Experience Replay},
  author = {Nan Jiang and Jinzhao Li and Yexiang Xue},
  journal= {arXiv preprint arXiv:2408.16999},
  year   = {2024}
}

Comments

This paper is accepted at RLC 2024