Cross-Silo Federated Learning: Challenges and Opportunities

TL;DR

This paper reviews cross-silo federated learning, highlighting its unique challenges, applications, and differences from cross-device FL, and discusses future research directions in this emerging field.

Contribution

It provides a systematic overview of cross-silo FL, emphasizing its challenges, applications, and distinctions from cross-device FL, and outlines future research directions.

Findings

Cross-silo FL involves organizational clients with fewer participants.

Major challenges include data heterogeneity and privacy concerns.

Future research needs to address scalability and security issues.

Abstract

Federated learning (FL) is an emerging technology that enables the training of machine learning models from multiple clients while keeping the data distributed and private. Based on the participating clients and the model training scale, federated learning can be classified into two types: cross-device FL where clients are typically mobile devices and the client number can reach up to a scale of millions; cross-silo FL where clients are organizations or companies and the client number is usually small (e.g., within a hundred). While existing studies mainly focus on cross-device FL, this paper aims to provide an overview of the cross-silo FL. More specifically, we first discuss applications of cross-silo FL and outline its major challenges. We then provide a systematic overview of the existing approaches to the challenges in cross-silo FL by focusing on their connections and differences to cross-device FL. Finally, we discuss future directions and open issues that merit research efforts from the community.