Internal Cross-layer Gradients for Extending Homogeneity to Heterogeneity in Federated Learning

TL;DR

This paper introduces InCo Aggregation, a novel method leveraging internal cross-layer gradients to extend the effectiveness of homogeneous federated learning algorithms in heterogeneous system environments, without extra communication overhead.

Contribution

The paper proposes a new training scheme using internal cross-layer gradients to improve model performance in heterogeneous federated learning settings, applicable to existing homogeneous methods.

Findings

InCo Aggregation improves performance in heterogeneous FL scenarios.

Higher layer similarities correlate with client performance.

Shallow layers exhibit higher similarity than deep layers.

Abstract

Federated learning (FL) inevitably confronts the challenge of system heterogeneity in practical scenarios. To enhance the capabilities of most model-homogeneous FL methods in handling system heterogeneity, we propose a training scheme that can extend their capabilities to cope with this challenge. In this paper, we commence our study with a detailed exploration of homogeneous and heterogeneous FL settings and discover three key observations: (1) a positive correlation between client performance and layer similarities, (2) higher similarities in the shallow layers in contrast to the deep layers, and (3) the smoother gradients distributions indicate the higher layer similarities. Building upon these observations, we propose InCo Aggregation that leverages internal cross-layer gradients, a mixture of gradients from shallow and deep layers within a server model, to augment the similarity in the deep layers without requiring additional communication between clients. Furthermore, our methods can be tailored to accommodate model-homogeneous FL methods such as FedAvg, FedProx, FedNova, Scaffold, and MOON, to expand their capabilities to handle the system heterogeneity. Copious experimental results validate the effectiveness of InCo Aggregation, spotlighting internal cross-layer gradients as a promising avenue to enhance the performance in heterogeneous FL.