FedADMM-InSa: An Inexact and Self-Adaptive ADMM for Federated Learning

Yongcun Song, Ziqi Wang, and Enrique Zuazua

Overview

This repository implements FedADMM-InSa, an inexact and self-adaptive ADMM method for federated learning. The training pipeline supports FedADMM-InSa and related federated optimization baselines.

Usage

Run a single FedADMM-InSa experiment using the YAML configuration:

python main.py --cfg utils/config.yaml

You can also override configuration values from the command line.

Citation

If this code is useful for your research, please cite the paper:

@article{swz2025fedadmminsa,
title = {FedADMM-InSa: An inexact and self-adaptive ADMM for federated learning},
journal = {Neural Networks},
volume = {181},
pages = {106772},
year = {2025},
issn = {0893-6080},
author = {Yongcun Song and Ziqi Wang and Enrique Zuazua}
}

Acknowledgments

Alphabetical authorship according to mathematical tradition. Funded by the European Union's Horizon Europe MSCA project ModConFlex (grant number 101073558)