Cross-Fusion Rule for Personalized Federated Learning

TL;DR

This paper introduces a multi-layer fusion framework for personalized federated learning that enhances collaboration efficiency and reduces overfitting by combining personalized and generic strategies based on network layer functions.

Contribution

It proposes a novel multi-layer fusion strategy with a new similarity metric and layer-specific fusion thresholds to improve personalized federated learning performance.

Findings

Outperforms state-of-the-art methods in experiments

Reduces overfitting in personalized models

Enhances collaboration efficiency among heterogeneous data

Abstract

Data scarcity and heterogeneity pose significant performance challenges for personalized federated learning, and these challenges are mainly reflected in overfitting and low precision in existing methods. To overcome these challenges, a multi-layer multi-fusion strategy framework is proposed in this paper, i.e., the server adopts the network layer parameters of each client upload model as the basic unit of fusion for information-sharing calculation. Then, a new fusion strategy combining personalized and generic is purposefully proposed, and the network layer number fusion threshold of each fusion strategy is designed according to the network layer function. Under this mechanism, the L2-Norm negative exponential similarity metric is employed to calculate the fusion weights of the corresponding feature extraction layer parameters for each client, thus improving the efficiency of heterogeneous data personalized collaboration. Meanwhile, the federated global optimal model approximation fusion strategy is adopted in the network full-connect layer, and this generic fusion strategy alleviates the overfitting introduced by forceful personalized. Finally, the experimental results show that the proposed method is superior to the state-of-the-art methods.