{"id":"arxiv/2610.34433","url":"https://arxiv.org/abs/2610.34433","tags":["math.OC","cs.LG","math.PR"],"score":0,"title":"Non-Ergodic Markovian Mirror Descent with Unbounded Noise","domain":"math.OC","source":"arxiv","authors":["Yunus Alp Biyikoglu","R. Srikant"],"reviews":0,"subject":"math.OC","summary":"We analyze stochastic mirror descent for minimizing a convex objective expressed as an expectation under the stationary distribution of a Markov chain, using noisy subgradient observations along a single sample path. Through Poisson's equation, we decompose the Markovian noise and control its contribution under regularity and moment conditions that allow the stochastic subgradients and the Poisson solution to be unbounded across the state space. We establish an $O(T^{-1/2})$ bound on the expected suboptimality gap of the weighted-average iterate for an appropriately chosen step size. The…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-25","reactions":{}}