{"id":"arxiv/2610.15202","url":"https://arxiv.org/abs/2610.15202","tags":["math.OC"],"score":0,"title":"Adaptive two-points zero-order accelerated methods for stochastic convex optimization problems","domain":"math.OC","source":"arxiv","authors":["Yuriy Dorn","Osman Osmanov","Timofei Loginov","Nazarii Tupitsa","Darina Dvinskikh","Alexander Gasnikov"],"reviews":0,"subject":"math.OC","summary":"We study stochastic convex optimization, where each sampled function can be evaluated at two arbitrary points. Our accelerated method chooses step sizes, batch sizes and smoothing radii from function values. A forward difference between smoothing scales controls the radius; we account for changes in the smoothed objective. Under the stated regularity and second-moment assumptions, exact function evaluations give \\begin{multline*} \\mathbb{E}\\bigl[f(x_N)-f(x^\\star)\\bigr] \\le \\widetilde{O}\\biggl( \\frac{\\sqrt{d}\\,M_2 D}{\\sqrt{T}} +\\min\\biggl( \\frac{L D^2}{N^2},\\; \\frac{d^{1/4} M…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-10","reactions":{}}