Reducing Training Time in Cross-Silo Federated Learning using Multigraph Topology

TL;DR

This paper introduces a multigraph topology for cross-silo federated learning that reduces training time by enabling asynchronous model aggregation through isolated nodes, without sacrificing model accuracy.

Contribution

The paper proposes a novel multigraph topology that improves training efficiency in cross-silo federated learning by allowing asynchronous updates via isolated nodes.

Findings

Significantly reduces training time compared to existing topologies.

Maintains comparable model accuracy with reduced training duration.

Validated on three public datasets with extensive experiments.

Abstract

Federated learning is an active research topic since it enables several participants to jointly train a model without sharing local data. Currently, cross-silo federated learning is a popular training setting that utilizes a few hundred reliable data silos with high-speed access links to training a model. While this approach has been widely applied in real-world scenarios, designing a robust topology to reduce the training time remains an open problem. In this paper, we present a new multigraph topology for cross-silo federated learning. We first construct the multigraph using the overlay graph. We then parse this multigraph into different simple graphs with isolated nodes. The existence of isolated nodes allows us to perform model aggregation without waiting for other nodes, hence effectively reducing the training time. Intensive experiments on three public datasets show that our proposed method significantly reduces the training time compared with recent state-of-the-art topologies while maintaining the accuracy of the learned model. Our code can be found at https://github.com/aioz-ai/MultigraphFL