Saved in:
Bibliographic Details
Main Authors: Guo, Wenyou, Qu, Ting, Pan, Chunrong, Huang, George Q.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.08606
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911240059617280
author Guo, Wenyou
Qu, Ting
Pan, Chunrong
Huang, George Q.
author_facet Guo, Wenyou
Qu, Ting
Pan, Chunrong
Huang, George Q.
contents Federated learning (FL), as a distributed collaborative machine learning (ML) framework under privacy-preserving constraints, has garnered increasing research attention in cross-organizational data collaboration scenarios. This paper proposes a class of distributed optimization algorithms based on the augmented Lagrangian technique, designed to accommodate diverse communication topologies in both centralized and decentralized FL settings. Furthermore, we develop multiple termination criteria and parameter update mechanisms to enhance computational efficiency, accompanied by rigorous theoretical guarantees of convergence. By generalizing the augmented Lagrangian relaxation through the incorporation of proximal relaxation and quadratic approximation, our framework systematically recovers a broad of classical unconstrained optimization methods, including proximal algorithm, classic gradient descent, and stochastic gradient descent, among others. Notably, the convergence properties of these methods can be naturally derived within the proposed theoretical framework. Numerical experiments demonstrate that the proposed algorithm exhibits strong performance in large-scale settings with significant statistical heterogeneity across clients.
format Preprint
id arxiv_https___arxiv_org_abs_2508_08606
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Distributed optimization: designed for federated learning
Guo, Wenyou
Qu, Ting
Pan, Chunrong
Huang, George Q.
Machine Learning
Optimization and Control
Federated learning (FL), as a distributed collaborative machine learning (ML) framework under privacy-preserving constraints, has garnered increasing research attention in cross-organizational data collaboration scenarios. This paper proposes a class of distributed optimization algorithms based on the augmented Lagrangian technique, designed to accommodate diverse communication topologies in both centralized and decentralized FL settings. Furthermore, we develop multiple termination criteria and parameter update mechanisms to enhance computational efficiency, accompanied by rigorous theoretical guarantees of convergence. By generalizing the augmented Lagrangian relaxation through the incorporation of proximal relaxation and quadratic approximation, our framework systematically recovers a broad of classical unconstrained optimization methods, including proximal algorithm, classic gradient descent, and stochastic gradient descent, among others. Notably, the convergence properties of these methods can be naturally derived within the proposed theoretical framework. Numerical experiments demonstrate that the proposed algorithm exhibits strong performance in large-scale settings with significant statistical heterogeneity across clients.
title Distributed optimization: designed for federated learning
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2508.08606