{"id":"arxiv/2604.16620","url":"https://arxiv.org/abs/2604.16620","tags":["cs.LG","math.OC"],"score":0,"title":"Lower Bounds and Proximally Anchored SGD for Non-Convex Minimization Under Unbounded Variance","domain":"cs.LG","source":"arxiv","authors":["Arda Fazla","Ege C. Kaya","Antesh Upadhyay","Abolfazl Hashemi"],"reviews":0,"subject":"cs.LG","summary":"Analysis of Stochastic Gradient Descent (SGD) and its variants typically relies on the assumption of uniformly bounded variance, a condition that frequently fails in practical non-convex settings, such as neural network training, as well as in several elementary optimization settings. While several relaxations are explored in the literature, the Blum-Gladyshev (BG-0) condition, which permits the variance to grow quadratically with distance has recently been shown to be the weakest condition. However, the study of the oracle complexity of stochastic first-order non-convex optimization under…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-04-17","reactions":{}}