FedPrism: Adaptive Personalized Federated Learning under Non-IID Data

TL;DR

FedPrism introduces an adaptive federated learning framework that decomposes client models into global, group, and private components, improving personalization and performance in highly non-IID data environments.

Contribution

The paper presents FedPrism, a novel framework combining Prism Decomposition and Dual-Stream design to enhance personalization in federated learning with non-IID data.

Findings

FedPrism outperforms static aggregation baselines in accuracy under high heterogeneity.

The framework effectively groups similar clients and adapts to data changes.

Experimental results show significant improvements in non-IID settings.

Abstract

Federated Learning (FL) suffers significant performance degradation in real-world deployments characterized by moderate to extreme statistical heterogeneity (non-IID client data). While global aggregation strategies promote broad generalization, they often fail to capture the diversity of local data distributions, leading to suboptimal personalization. We address this problem with FedPrism, a framework that uses two main strategies. First, it uses a Prism Decomposition method that builds each client's model from three parts: a global foundation, a shared group part for similar clients, and a private part for unique local data. This allows the system to group similar users together automatically and adapt if their data changes. Second, we include a Dual-Stream design that runs a general model alongside a local specialist. The system routes predictions between the general model and the local specialist based on the specialist's confidence. Through systematic experiments on non-IID data partitions, we demonstrate that FedPrism exceeds static aggregation and hard-clustering baselines, achieving significant accuracy gains under high heterogeneity. These results establish FedPrism as a robust and flexible solution for federated learning in heterogeneous environments, effectively balancing generalizable knowledge with adaptive personalization.