Differentiated Aggregation to Improve Generalization in Federated Learning

arXiv:2404.11754v4 Announce Type: replace-cross Abstract: This paper focuses on reducing the communication cost of federated learning by exploring generalization bounds and representation learning. We first characterize a tighter generalization bound for one-round federated learning based on local clients' generalizations and heterogeneity of data distribution (non-iid scenario). We also characterize a generalization bound in R-round federated learning and its relation to the number of local updates (local stochastic gradient descents (SGDs)). Then, based on our generalization bound analysis and its interpretation through representation learning, we infer that less frequent aggregations for the representation extractor (typically corresponds to initial layers) compared to the head (usually the final layers) leads to the creation of more generalizable models, particularly in non-iid scenarios. We design a novel Federated Learning with Adaptive Local Steps (FedALS) algorithm based on our generalization bound and representation learning analysis. FedALS employs varying aggregation frequencies for different parts of the model, so reduces the communication cost. The paper is followed with experimental results showing the effectiveness of FedALS. Our codes are available at for reproducibility.

Abstract:This paper focuses on reducing the communication cost of federated learning by exploring generalization bounds and representation learning. We first characterize a tighter generalization bound for one-round federated learning based on local clients' generalizations and heterogeneity of data distribution (non-iid scenario). We also characterize a generalization bound in R-round federated learning and its relation to the number of local updates (local stochastic gradient descents (SGDs)). Then, based on our generalization bound analysis and its interpretation through representation learning, we infer that less frequent aggregations for the representation extractor (typically corresponds to initial layers) compared to the head (usually the final layers) leads to the creation of more generalizable models, particularly in non-iid scenarios. We design a novel Federated Learning with Adaptive Local Steps (FedALS) algorithm based on our generalization bound and representation learning analysis. FedALS employs varying aggregation frequencies for different parts of the model, so reduces the communication cost. The paper is followed with experimental results showing the effectiveness of FedALS. Our codes are available at for reproducibility.

25 pages. Published in Transactions on Machine Learning Research (TMLR), 2025

Machine Learning (cs.LG); Artificial Intelligence (cs.AI)

arXiv:2404.11754 [cs.LG]

(or arXiv:2404.11754v4 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2404.11754 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

analiza aiqbench

Differentiated Aggregation to Improve Generalization in Federated Learning

Subiecte conexe