English

Regularized Q-learning through Robust Averaging

Optimization and Control 2024-05-30 v2 Machine Learning

Abstract

We propose a new Q-learning variant, called 2RA Q-learning, that addresses some weaknesses of existing Q-learning methods in a principled manner. One such weakness is an underlying estimation bias which cannot be controlled and often results in poor performance. We propose a distributionally robust estimator for the maximum expected value term, which allows us to precisely control the level of estimation bias introduced. The distributionally robust estimator admits a closed-form solution such that the proposed algorithm has a computational cost per iteration comparable to Watkins' Q-learning. For the tabular case, we show that 2RA Q-learning converges to the optimal policy and analyze its asymptotic mean-squared error. Lastly, we conduct numerical experiments for various settings, which corroborate our theoretical findings and indicate that 2RA Q-learning often performs better than existing methods.

Keywords

Cite

@article{arxiv.2405.02201,
  title  = {Regularized Q-learning through Robust Averaging},
  author = {Peter Schmitt-Förster and Tobias Sutter},
  journal= {arXiv preprint arXiv:2405.02201},
  year   = {2024}
}

Comments

26 pages, 5 figures