{"id":"arxiv/2603.07149","url":"https://arxiv.org/abs/2603.07149","tags":["math.PR"],"score":0,"title":"Quantitative Fluctuation Analysis for Continuous-Time Stochastic Gradient Descent via Malliavin Calculus","domain":"math.PR","source":"arxiv","authors":["Solesne Bourguin","Shivam Singh Dhama","Konstantinos Spiliopoulos"],"reviews":0,"subject":"math.PR","summary":"In this paper, we establish a Quantitative Central Limit Theorem (QCLT) for the Stochastic Gradient Descent in Continuous Time (SGDCT) algorithm, whose parameter updates are governed by a stochastic differential equation. We derive an explicit rate at which the SGDCT iterates converge, in the Wasserstein metric, to a critical point of the objective function. This rate is driven primarily by the magnitude of the learning rate: for a fixed convexity constant of the objective function, smaller learning rates lead to slower convergence. Our approach relies on tools from Malliavin calculus. In…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-07","reactions":{}}