A General Continuous-Time Formulation of Stochastic ADMM and Its Variants

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Li, Chris Junchi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913324245975040
author Li, Chris Junchi
author_facet Li, Chris Junchi
contents Stochastic versions of the alternating direction method of multiplier (ADMM) and its variants play a key role in many modern large-scale machine learning problems. In this work, we introduce a unified algorithmic framework called generalized stochastic ADMM and investigate their continuous-time analysis. The generalized framework widely includes many stochastic ADMM variants such as standard, linearized and gradient-based ADMM. Our continuous-time analysis provides us with new insights into stochastic ADMM and variants, and we rigorously prove that under some proper scaling, the trajectory of stochastic ADMM weakly converges to the solution of a stochastic differential equation with small noise. Our analysis also provides a theoretical explanation of why the relaxation parameter should be chosen between 0 and 2.
format Preprint
id arxiv_https___arxiv_org_abs_2404_14358
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A General Continuous-Time Formulation of Stochastic ADMM and Its Variants
Li, Chris Junchi
Optimization and Control
Machine Learning
Stochastic versions of the alternating direction method of multiplier (ADMM) and its variants play a key role in many modern large-scale machine learning problems. In this work, we introduce a unified algorithmic framework called generalized stochastic ADMM and investigate their continuous-time analysis. The generalized framework widely includes many stochastic ADMM variants such as standard, linearized and gradient-based ADMM. Our continuous-time analysis provides us with new insights into stochastic ADMM and variants, and we rigorously prove that under some proper scaling, the trajectory of stochastic ADMM weakly converges to the solution of a stochastic differential equation with small noise. Our analysis also provides a theoretical explanation of why the relaxation parameter should be chosen between 0 and 2.
title A General Continuous-Time Formulation of Stochastic ADMM and Its Variants
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2404.14358