Personalized Federated Learning for Heterogeneous Clients with Clustered Knowledge Transfer
This addresses the challenge of realistic federated learning scenarios where clients have diverse capabilities and constraints, offering a more feasible solution for personalized model training.
The paper tackles the problem of personalized federated learning for clients with heterogeneous data, system capabilities, and limited communication resources by proposing PerFed-CKT, a framework that allows heterogeneous model architectures and uses clustered co-distillation with logits for knowledge transfer, achieving high test accuracy with several orders of magnitude lower communication cost compared to state-of-the-art methods.
Personalized federated learning (FL) aims to train model(s) that can perform well for individual clients that are highly data and system heterogeneous. Most work in personalized FL, however, assumes using the same model architecture at all clients and increases the communication cost by sending/receiving models. This may not be feasible for realistic scenarios of FL. In practice, clients have highly heterogeneous system-capabilities and limited communication resources. In our work, we propose a personalized FL framework, PerFed-CKT, where clients can use heterogeneous model architectures and do not directly communicate their model parameters. PerFed-CKT uses clustered co-distillation, where clients use logits to transfer their knowledge to other clients that have similar data-distributions. We theoretically show the convergence and generalization properties of PerFed-CKT and empirically show that PerFed-CKT achieves high test accuracy with several orders of magnitude lower communication cost compared to the state-of-the-art personalized FL schemes.