A Coalition Formation Game Approach for Personalized Federated Learning
This addresses the problem of improving personalized model performance in federated learning for clients with diverse data distributions, representing an incremental advance over existing pairwise collaboration methods.
The paper tackles statistical diversity in personalized federated learning by proposing pFedSV, a method that uses Shapley value from coalition game theory to optimize multiwise client collaborations, achieving superior personalized accuracy on datasets like MNIST, Fashion-MNIST, and CIFAR-10 under Non-IID settings.
Facing the challenge of statistical diversity in client local data distribution, personalized federated learning (PFL) has become a growing research hotspot. Although the state-of-the-art methods with model similarity-based pairwise collaboration have achieved promising performance, they neglect the fact that model aggregation is essentially a collaboration process within the coalition, where the complex multiwise influences take place among clients. In this paper, we first apply Shapley value (SV) from coalition game theory into the PFL scenario. To measure the multiwise collaboration among a group of clients on the personalized learning performance, SV takes their marginal contribution to the final result as a metric. We propose a novel personalized algorithm: pFedSV, which can 1. identify each client's optimal collaborator coalition and 2. perform personalized model aggregation based on SV. Extensive experiments on various datasets (MNIST, Fashion-MNIST, and CIFAR-10) are conducted with different Non-IID data settings (Pathological and Dirichlet). The results show that pFedSV can achieve superior personalized accuracy for each client, compared to the state-of-the-art benchmarks.