Stochastic Modified Flows for Riemannian Stochastic Gradient Descent

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Gess, Benjamin, Kassing, Sebastian, Rana, Nimit
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916646008913920
author Gess, Benjamin
Kassing, Sebastian
Rana, Nimit
author_facet Gess, Benjamin
Kassing, Sebastian
Rana, Nimit
contents We give quantitative estimates for the rate of convergence of Riemannian stochastic gradient descent (RSGD) to Riemannian gradient flow and to a diffusion process, the so-called Riemannian stochastic modified flow (RSMF). Using tools from stochastic differential geometry we show that, in the small learning rate regime, RSGD can be approximated by the solution to the RSMF driven by an infinite-dimensional Wiener process. The RSMF accounts for the random fluctuations of RSGD and, thereby, increases the order of approximation compared to the deterministic Riemannian gradient flow. The RSGD is build using the concept of a retraction map, that is, a cost efficient approximation of the exponential map, and we prove quantitative bounds for the weak error of the diffusion approximation under assumptions on the retraction map, the geometry of the manifold, and the random estimators of the gradient.
format Preprint
id arxiv_https___arxiv_org_abs_2402_03467
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Stochastic Modified Flows for Riemannian Stochastic Gradient Descent
Gess, Benjamin
Kassing, Sebastian
Rana, Nimit
Machine Learning
Optimization and Control
Probability
Primary 62L20, Secondary 58J65, 60J20, 65K05
We give quantitative estimates for the rate of convergence of Riemannian stochastic gradient descent (RSGD) to Riemannian gradient flow and to a diffusion process, the so-called Riemannian stochastic modified flow (RSMF). Using tools from stochastic differential geometry we show that, in the small learning rate regime, RSGD can be approximated by the solution to the RSMF driven by an infinite-dimensional Wiener process. The RSMF accounts for the random fluctuations of RSGD and, thereby, increases the order of approximation compared to the deterministic Riemannian gradient flow. The RSGD is build using the concept of a retraction map, that is, a cost efficient approximation of the exponential map, and we prove quantitative bounds for the weak error of the diffusion approximation under assumptions on the retraction map, the geometry of the manifold, and the random estimators of the gradient.
title Stochastic Modified Flows for Riemannian Stochastic Gradient Descent
topic Machine Learning
Optimization and Control
Probability
Primary 62L20, Secondary 58J65, 60J20, 65K05
url https://arxiv.org/abs/2402.03467