HADFL: Heterogeneity-aware Decentralized Federated Learning Framework

TL;DR

HADFL is a decentralized, asynchronous federated learning framework that efficiently handles device heterogeneity, reduces communication overhead, and achieves significant speedups without sacrificing model accuracy.

Contribution

HADFL introduces a heterogeneity-aware, decentralized asynchronous training framework that improves efficiency and speedup in federated learning on heterogeneous devices.

Findings

Achieves up to 3.15x speedup over decentralized-FedAvg.

Achieves up to 4.68x speedup over Pytorch distributed training.

Maintains comparable convergence accuracy.

Abstract

Federated learning (FL) supports training models on geographically distributed devices. However, traditional FL systems adopt a centralized synchronous strategy, putting high communication pressure and model generalization challenge. Existing optimizations on FL either fail to speedup training on heterogeneous devices or suffer from poor communication efficiency. In this paper, we propose HADFL, a framework that supports decentralized asynchronous training on heterogeneous devices. The devices train model locally with heterogeneity-aware local steps using local data. In each aggregation cycle, they are selected based on probability to perform model synchronization and aggregation. Compared with the traditional FL system, HADFL can relieve the central server's communication pressure, efficiently utilize heterogeneous computing power, and can achieve a maximum speedup of 3.15x than decentralized-FedAvg and 4.68x than Pytorch distributed training scheme, respectively, with almost no loss of convergence accuracy.