Tram-FL: Routing-based Model Training for Decentralized Federated Learning

TL;DR

Tram-FL introduces a routing-based decentralized federated learning method that refines a global model through sequential node transfers, reducing communication and improving accuracy in non-IID data scenarios.

Contribution

It proposes a novel routing algorithm and a sequential model transfer approach, enhancing model accuracy and communication efficiency in decentralized federated learning.

Findings

Outperforms baseline methods on MNIST, CIFAR-10, and IMDb datasets.

Reduces communication costs compared to traditional federated learning.

Achieves high accuracy under non-IID data distributions.

Abstract

In decentralized federated learning (DFL), substantial traffic from frequent inter-node communication and non-independent and identically distributed (non-IID) data challenges high-accuracy model acquisition. We propose Tram-FL, a novel DFL method, which progressively refines a global model by transferring it sequentially amongst nodes, rather than by exchanging and aggregating local models. We also introduce a dynamic model routing algorithm for optimal route selection, aimed at enhancing model precision with minimal forwarding. Our experiments using MNIST, CIFAR-10, and IMDb datasets demonstrate that Tram-FL with the proposed routing delivers high model accuracy under non-IID conditions, outperforming baselines while reducing communication costs.