Arnaud Guillin, Yu Wang, Lihu Xu, Haoran Yang
Published 2024-10-22Version 1
Stochastic gradient descent with momentum is a popular variant of stochastic gradient descent, which has recently been reported to have a close relationship with the underdamped Langevin diffusion. In this paper, we establish a quantitative error estimate between them in the 1-Wasserstein and total variation distances.
The convergence of the Stochastic Gradient Descent (SGD) : a self-contained proof
Gaussian Approximation and Multiplier Bootstrap for Stochastic Gradient Descent
Algorithmic Stability of Stochastic Gradient Descent with Momentum under Heavy-Tailed Noise
![QR code for arXiv:2410.17297 [stat.ML]](http://oss.arxitics.com/images/favicon.svg)