GPFL: A Gradient Projection-Based Client Selection Framework for Efficient Federated Learning
This work addresses client selection challenges in federated learning for applications with data heterogeneity, offering incremental improvements over existing methods.
The paper tackled the problem of client selection in federated learning to balance model accuracy and communication efficiency, proposing GPFL, which improved test accuracy by over 9% on the FEMINST dataset in Non-IID scenarios and reduced computation times.
Federated learning client selection is crucial for determining participant clients while balancing model accuracy and communication efficiency. Existing methods have limitations in handling data heterogeneity, computational burdens, and independent client treatment. To address these challenges, we propose GPFL, which measures client value by comparing local and global descent directions. We also employ an Exploit-Explore mechanism to enhance performance. Experimental results on FEMINST and CIFAR-10 datasets demonstrate that GPFL outperforms baselines in Non-IID scenarios, achieving over 9\% improvement in FEMINST test accuracy. Moreover, GPFL exhibits shorter computation times through pre-selection and parameter reuse in federated learning.