Mixed Federated Learning: Joint Decentralized and Centralized Learning

TL;DR

This paper introduces mixed federated learning, combining decentralized and centralized training with new algorithms, enabling better data utilization and reducing communication and computation costs while maintaining privacy.

Contribution

It proposes the concept of mixed FL, introduces three algorithms with convergence analysis, and demonstrates significant efficiency gains through extensive experiments.

Findings

Mixed FL can match oracle accuracy on inference data.

Reduces communication and computation overhead by over 90%.

Algorithms perform as predicted by theoretical bounds.

Abstract

Federated learning (FL) enables learning from decentralized privacy-sensitive data, with computations on raw data confined to take place at edge clients. This paper introduces mixed FL, which incorporates an additional loss term calculated at the coordinating server (while maintaining FL's private data restrictions). There are numerous benefits. For example, additional datacenter data can be leveraged to jointly learn from centralized (datacenter) and decentralized (federated) training data and better match an expected inference data distribution. Mixed FL also enables offloading some intensive computations (e.g., embedding regularization) to the server, greatly reducing communication and client computation load. For these and other mixed FL use cases, we present three algorithms: PARALLEL TRAINING, 1-WAY GRADIENT TRANSFER, and 2-WAY GRADIENT TRANSFER. We state convergence bounds for each, and give intuition on which are suited to particular mixed FL problems. Finally we perform extensive experiments on three tasks, demonstrating that mixed FL can blend training data to achieve an oracle's accuracy on an inference distribution, and can reduce communication and computation overhead by over 90%. Our experiments confirm theoretical predictions of how algorithms perform under different mixed FL problem settings.