arXiv:2602.20403 · v3

Wasserstein Distributionally Robust Online Learning

首次提交 2026-02-24 06:55

版本更新 2026-10-28 03:48

雷达观测 2026-10-30 10:23

摘要

We study online Wasserstein distributionally robust optimization (DRO), where the ambiguity set is a Wasserstein ball centered at the empirical distribution and evolves with each observation. We formulate each round as a zero-sum game and develop an online distributional best-response algorithm that requires only one projected subgradient step per iteration. Combining regret analysis with measure concentration, we establish explicit, instance-dependent, and high-probability convergence rates to the optimal offline DRO value.