Abstract:We study the problem of learning Nash equilibria in offline two-player zero-sum Markov games. While existing approaches often rely on explicit pessimism to address distribution shift, we show that KL regularization alone suffices to stabilize learning and guarantee convergence. We first introduce Regularized Offline Sequential Equilibrium (ROSE), a theoretical framework that achieves a fast $\widetilde{\mathcal{O}}(1/n)$ convergence rate under \textit{unilateral concentrability}, improving over the standard $\widetilde{\mathcal{O}}(1/\sqrt{n})$ rates in unregularized settings. We then propose Sequential Offline Self-play Mirror Descent (SOS-MD), a practical model-free algorithm based on least-squares value estimation and iterative self-play updates. We prove that the last iterate of SOS-MD attains the same $\widetilde{\mathcal{O}}(1/n)$ statistical rate up to a vanishing optimization error of order $\widetilde{\mathcal{O}}(1/\sqrt{T})$ in the number of self-play iterations $T$.
| Subjects: | Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT) |
| Cite as: | arXiv:2605.13025 [cs.LG] |
| (or arXiv:2605.13025v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.13025 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Claire Chen [view email]
[v1]
Wed, 13 May 2026 05:29:21 UTC (160 KB)