Stochastic Modified Flows, Mean-Field Limits and Dynamics of Stochastic Gradient Descent
Benjamin Gess, Sebastian Kassing, Vitalii Konarovskyi.
Year: 2024, Volume: 25, Issue: 30, Pages: 1−27
Abstract
We propose new limiting dynamics for stochastic gradient descent in the small learning rate regime called stochastic modified flows. These SDEs are driven by a cylindrical Brownian motion and improve the so-called stochastic modified equations by having regular diffusion coefficients and by matching the multi-point statistics. As a second contribution, we introduce distribution dependent stochastic modified flows which we prove to describe the fluctuating limiting dynamics of stochastic gradient descent in the small learning rate - infinite width scaling regime.