Blockchain-aided wireless federated learning: Resource allocation and client scheduling

TL;DR

This paper proposes a blockchain-aided decentralized federated learning framework for wireless networks, optimizing resource allocation and client scheduling to reduce training delay while maintaining accuracy.

Contribution

It introduces the DRC-BDFL algorithm that jointly optimizes resource use and client participation using Lyapunov optimization, addressing wireless network constraints.

Findings

DRC-BDFL reduces training delay by over 9% on SVHN and CIFAR-10 datasets.

The proposed method maintains comparable accuracy to baseline algorithms.

An upper bound for convergence of the learning process is derived.

Abstract

Federated learning (FL) based on the centralized design faces both challenges regarding the trust issue and a single point of failure. To alleviate these issues, blockchain-aided decentralized FL (BDFL) introduces the decentralized network architecture into the FL training process, which can effectively overcome the defects of centralized architecture. However, deploying BDFL in wireless networks usually encounters challenges such as limited bandwidth, computing power, and energy consumption. Driven by these considerations, a dynamic stochastic optimization problem is formulated to minimize the average training delay by jointly optimizing the resource allocation and client selection under the constraints of limited energy budget and client participation. We solve the long-term mixed integer non-linear programming problem by employing the tool of Lyapunov optimization and thereby propose the dynamic resource allocation and client scheduling BDFL (DRC-BDFL) algorithm. Furthermore, we analyze the learning performance of DRC-BDFL and derive an upper bound for convergence regarding the global loss function. Extensive experiments conducted on SVHN and CIFAR-10 datasets demonstrate that DRC-BDFL achieves comparable accuracy to baseline algorithms while significantly reducing the training delay by 9.24% and 12.47%, respectively.