Contribution Evaluation in Federated Learning: Examining Current Approaches
This work addresses the need for fair and efficient CE methods to support the mainstream adoption of Federated Learning, but it is incremental as it reviews and benchmarks existing approaches with a new addition.
The paper tackles the problem of Contribution Evaluation (CE) in Federated Learning, which involves quantifying the worth of client contributions based on local model updates, and benchmarks state-of-the-art approaches along with a new method on MNIST and CIFAR-10 datasets.
Federated Learning (FL) has seen increasing interest in cases where entities want to collaboratively train models while maintaining privacy and governance over their data. In FL, clients with private and potentially heterogeneous data and compute resources come together to train a common model without raw data ever leaving their locale. Instead, the participants contribute by sharing local model updates, which, naturally, differ in quality. Quantitatively evaluating the worth of these contributions is termed the Contribution Evaluation (CE) problem. We review current CE approaches from the underlying mathematical framework to efficiently calculate a fair value for each client. Furthermore, we benchmark some of the most promising state-of-the-art approaches, along with a new one we introduce, on MNIST and CIFAR-10, to showcase their differences. Designing a fair and efficient CE method, while a small part of the overall FL system design, is tantamount to the mainstream adoption of FL.