{"id":"arxiv/2603.01951","url":"https://arxiv.org/abs/2603.01951","tags":["cs.LG","math.OC","stat.ML"],"score":0,"title":"Accelerating Single-Pass SGD for Generalized Linear Prediction","domain":"cs.LG","source":"arxiv","authors":["Qian Chen","Shihong Ding","Cong Fang"],"reviews":0,"subject":"cs.LG","summary":"We study generalized linear prediction under a streaming setting, where each iteration uses only one fresh data point for a gradient-level update. While momentum is well-established in deterministic optimization, a fundamental open question is whether it can accelerate such single-pass non-quadratic stochastic optimization. We propose the first algorithm that successfully incorporates momentum via a novel data-dependent proximal method, achieving dual-momentum acceleration. Our derived excess risk bound decomposes into three components: an improved optimization error, a minimax optimal…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-03-02","reactions":{}}