English

The Plug-in Approach for Average-Reward and Discounted MDPs: Optimal Sample Complexity Analysis

Machine Learning 2025-02-12 v2 Information Theory math.IT Optimization and Control Machine Learning

Abstract

We study the sample complexity of the plug-in approach for learning ε\varepsilon-optimal policies in average-reward Markov decision processes (MDPs) with a generative model. The plug-in approach constructs a model estimate then computes an average-reward optimal policy in the estimated model. Despite representing arguably the simplest algorithm for this problem, the plug-in approach has never been theoretically analyzed. Unlike the more well-studied discounted MDP reduction method, the plug-in approach requires no prior problem information or parameter tuning. Our results fill this gap and address the limitations of prior approaches, as we show that the plug-in approach is optimal in several well-studied settings without using prior knowledge. Specifically it achieves the optimal diameter- and mixing-based sample complexities of O~(SADε2)\widetilde{O}\left(SA \frac{D}{\varepsilon^2}\right) and O~(SAτunifε2)\widetilde{O}\left(SA \frac{\tau_{\mathrm{unif}}}{\varepsilon^2}\right), respectively, without knowledge of the diameter DD or uniform mixing time τunif\tau_{\mathrm{unif}}. We also obtain span-based bounds for the plug-in approach, and complement them with algorithm-specific lower bounds suggesting that they are unimprovable. Our results require novel techniques for analyzing long-horizon problems which may be broadly useful and which also improve results for the discounted plug-in approach, removing effective-horizon-related sample size restrictions and obtaining the first optimal complexity bounds for the full range of sample sizes without reward perturbation.

Keywords

Cite

@article{arxiv.2410.07616,
  title  = {The Plug-in Approach for Average-Reward and Discounted MDPs: Optimal Sample Complexity Analysis},
  author = {Matthew Zurek and Yudong Chen},
  journal= {arXiv preprint arXiv:2410.07616},
  year   = {2025}
}

Comments

Accepted to 36th International Conference on Algorithmic Learning Theory (ALT 2025)

R2 v1 2026-06-28T19:15:38.314Z