FedECADO: A Dynamical System Model of Federated Learning

Aayushya Agarwal, Gauri Joshi, Lawrence Pileggi
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:531-549, 2025.

Abstract

Federated learning harnesses the power of distributed optimization to train a unified machine learning model across separate clients. However, heterogeneous data distributions and computational workloads can lead to inconsistent updates and limit model performance. This work tackles these challenges by proposing FedECADO, a new algorithm inspired by a dynamical system representation of the federated learning process. FedECADO addresses non-IID data distribution through an aggregate sensitivity model that reflects the amount of data processed by each client. To tackle heterogeneous computing, we design a multi-rate integration method with adaptive step-size selections that synchronizes active client updates in continuous time. Compared to prominent techniques, including FedProx, FedExp, and FedNova, FedECADO achieves higher classification accuracies in numerous heterogeneous scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-agarwal25c, title = {{F}ed{ECADO}: A Dynamical System Model of Federated Learning}, author = {Agarwal, Aayushya and Joshi, Gauri and Pileggi, Lawrence}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {531--549}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/agarwal25c/agarwal25c.pdf}, url = {https://proceedings.mlr.press/v267/agarwal25c.html}, abstract = {Federated learning harnesses the power of distributed optimization to train a unified machine learning model across separate clients. However, heterogeneous data distributions and computational workloads can lead to inconsistent updates and limit model performance. This work tackles these challenges by proposing FedECADO, a new algorithm inspired by a dynamical system representation of the federated learning process. FedECADO addresses non-IID data distribution through an aggregate sensitivity model that reflects the amount of data processed by each client. To tackle heterogeneous computing, we design a multi-rate integration method with adaptive step-size selections that synchronizes active client updates in continuous time. Compared to prominent techniques, including FedProx, FedExp, and FedNova, FedECADO achieves higher classification accuracies in numerous heterogeneous scenarios.} }
Endnote
%0 Conference Paper %T FedECADO: A Dynamical System Model of Federated Learning %A Aayushya Agarwal %A Gauri Joshi %A Lawrence Pileggi %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-agarwal25c %I PMLR %P 531--549 %U https://proceedings.mlr.press/v267/agarwal25c.html %V 267 %X Federated learning harnesses the power of distributed optimization to train a unified machine learning model across separate clients. However, heterogeneous data distributions and computational workloads can lead to inconsistent updates and limit model performance. This work tackles these challenges by proposing FedECADO, a new algorithm inspired by a dynamical system representation of the federated learning process. FedECADO addresses non-IID data distribution through an aggregate sensitivity model that reflects the amount of data processed by each client. To tackle heterogeneous computing, we design a multi-rate integration method with adaptive step-size selections that synchronizes active client updates in continuous time. Compared to prominent techniques, including FedProx, FedExp, and FedNova, FedECADO achieves higher classification accuracies in numerous heterogeneous scenarios.
APA
Agarwal, A., Joshi, G. & Pileggi, L.. (2025). FedECADO: A Dynamical System Model of Federated Learning. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:531-549 Available from https://proceedings.mlr.press/v267/agarwal25c.html.

Related Material