Unlocking the Potential of Weighting Methods in Federated Learning Through Communication Compression
Abstract
Modern machine learning problems are frequently formulated in federated learning domain and incorporate inherently heterogeneous data. Weighting methods operate efficiently in terms of iteration complexity and represent a common direction in this setting. At the same time, they do not address directly the main obstacle in federated and distributed learning -- communication bottleneck. We tackle this issue by incorporating compression into the weighting scheme. We establish the convergence under a convexity assumption, considering both exact and stochastic oracles. Finally, we evaluate the practical performance of the proposed method on real-world problems.
Cite
Text
Parfenov et al. "Unlocking the Potential of Weighting Methods in Federated Learning Through Communication Compression." International Conference on Learning Representations, 2026.Markdown
[Parfenov et al. "Unlocking the Potential of Weighting Methods in Federated Learning Through Communication Compression." International Conference on Learning Representations, 2026.](https://mlanthology.org/iclr/2026/parfenov2026iclr-unlocking/)BibTeX
@inproceedings{parfenov2026iclr-unlocking,
title = {{Unlocking the Potential of Weighting Methods in Federated Learning Through Communication Compression}},
author = {Parfenov, Valery and Bashirov, Nail and Medyakov, Daniil and Bylinkin, Dmitry and Beznosikov, Aleksandr},
booktitle = {International Conference on Learning Representations},
year = {2026},
url = {https://mlanthology.org/iclr/2026/parfenov2026iclr-unlocking/}
}