Fast-Convergent Federated Learning

Understand

Federated learning has emerged recently as a promising solution for distributing machine learning tasks through modern networks of mobile devices.

• Recent studies have obtained lower bounds on the expected decrease in model loss that is achieved through each round of federated learning.
• However, convergence generally requires a large number of communication rounds, which induces delay in model training and is costly in terms of network resources.
• In this paper, we propose a fast-convergent federated learning algorithm, called FOLB, which performs intelligent sampling of devices in each round of model training to optimize the expected convergence speed.