×

FedML:
A Research Library and Benchmark for
Federated Machine Learning
Federated learning is a fast-growing research field in distributed machine learning. Despite considerable research efforts have been made, there is no standard library and benchmark that supports fast developing FL algorithms and fair comparison of the performance of these FL algorithms. In this work, we introduce FedML, an open research library and benchmark that facilitates the development of new FL algorithms and fair performance comparison with existing algorithms under a variety of commonly used FL settings. We believe FedML will provide an efficient and reproducible means of developing and evaluating federated learning algorithms to the FL research community.
FedML:
A Research Library and Benchmark for
Federated Machine Learning

Federated learning is a fast-growing research field in distributed machine learning. Despite considerable research efforts have been made, there is no standard library and benchmark that supports fast developing FL algorithms and fair comparison of the performance of these FL algorithms. In this work, we introduce FedML, an open research library and benchmark that facilitates the development of new FL algorithms and fair performance comparison with existing algorithms under a variety of commonly used FL settings. We believe FedML will provide an efficient and reproducible means of developing and evaluating federated learning algorithms to the FL research community.