Modular Distributed Nonconvex Learning with Error Feedback
Optimization and Control
2025-07-01 v2 Machine Learning
Systems and Control
Systems and Control
Abstract
In this paper, we design a novel distributed learning algorithm using stochastic compressed communications. In detail, we pursue a modular approach, merging ADMM and a gradient-based approach, benefiting from the robustness of the former and the computational efficiency of the latter. Additionally, we integrate a stochastic integral action (error feedback) enabling almost sure rejection of the compression error. We analyze the resulting method in nonconvex scenarios and guarantee almost sure asymptotic convergence to the set of stationary points of the problem. This result is obtained using system-theoretic tools based on stochastic timescale separation. We corroborate our findings with numerical simulations in nonconvex classification.
Cite
@article{arxiv.2503.14055,
title = {Modular Distributed Nonconvex Learning with Error Feedback},
author = {Guido Carnevale and Nicola Bastianello},
journal= {arXiv preprint arXiv:2503.14055},
year = {2025}
}