CDKT-FL: Cross-Device Knowledge Transfer using Proxy Dataset in Federated Learning
This work addresses the problem of improving federated learning efficiency and stability for applications with heterogeneous client data, though it appears incremental as it builds on existing knowledge transfer mechanisms.
The paper tackles the challenge of robust generalization and personalization in federated learning under non-i.i.d. and small data conditions by proposing a knowledge distillation-based method for cross-device knowledge transfer using a proxy dataset. It achieves significant speedups and high personalized performance on local models across three federated datasets, with minimal communication load.
In a practical setting, how to enable robust Federated Learning (FL) systems, both in terms of generalization and personalization abilities, is one important research question. It is a challenging issue due to the consequences of non-i.i.d. properties of client's data, often referred to as statistical heterogeneity, and small local data samples from the various data distributions. Therefore, to develop robust generalized global and personalized models, conventional FL methods need to redesign the knowledge aggregation from biased local models while considering huge divergence of learning parameters due to skewed client data. In this work, we demonstrate that the knowledge transfer mechanism achieves these objectives and develop a novel knowledge distillation-based approach to study the extent of knowledge transfer between the global model and local models. Henceforth, our method considers the suitability of transferring the outcome distribution and (or) the embedding vector of representation from trained models during cross-device knowledge transfer using a small proxy dataset in heterogeneous FL. In doing so, we alternatively perform cross-device knowledge transfer following general formulations as 1) global knowledge transfer and 2) on-device knowledge transfer. Through simulations on three federated datasets, we show the proposed method achieves significant speedups and high personalized performance of local models. Furthermore, the proposed approach offers a more stable algorithm than other baselines during the training, with minimal communication data load when exchanging the trained model's outcomes and representation.