{"id":"arxiv/2602.05489","url":"https://arxiv.org/abs/2602.05489","tags":["math.OC","cs.LG","stat.ML"],"score":0,"title":"Convergence Rate of the Last Iterate of Stochastic Proximal Algorithms","domain":"math.OC","source":"arxiv","authors":["Kevin Kurian Thomas Vaidyan","Michael P. Friedlander","Ahmet Alacaoglu"],"reviews":0,"subject":"math.OC","summary":"We analyze two classical algorithms for solving additively composite convex optimization problems where the objective is the sum of a smooth term and a nonsmooth regularizer: proximal stochastic gradient method for a single regularizer; and the randomized incremental proximal method, which uses the proximal operator of a randomly selected function when the regularizer is given as the sum of many nonsmooth functions. We focus on relaxing the bounded variance assumption that is common, yet stringent, for getting last iterate convergence rates. We prove the $\\widetilde{O}(1/\\sqrt{T})$ rate of…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-05","reactions":{}}