{"id":"arxiv/2611.13383","url":"https://arxiv.org/abs/2611.13383","tags":["cs.LG","math.OC","stat.ML"],"score":0,"title":"Settling the Burn-in Gap: Sharp Risk Bounds for Tail-Averaged SGD in Linear Regression","domain":"cs.LG","source":"arxiv","authors":["Qingyue Zhao","Xuheng Li","Quanquan Gu"],"reviews":0,"subject":"cs.LG","summary":"We study the excess risk of constant-stepsize online stochastic gradient descent (SGD) with tail-averaging for high-dimensional linear regression. Building on the instance-dependent spectral framework of Zou et al. (2021), we establish matching initialization-explicit upper and lower bounds on the total excess risk of tail-averaged SGD in terms of the population covariance spectrum $\\mathbf{H}$ and initialization $\\mathbf{w}_0-\\mathbf{w}^*$ whenever a trajectory spectral feedback condition holds. Existing analyses exhibit a fundamental gap in the effective variance: the upper bound of Zou et…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-11-10","reactions":{}}