MPFL: A Decentralised Federated Learning Framework Based on Multi-Population Genetic Algorithm
Wenqi Ding, Yuanchao Liu, Zhongjie Wang, Zheng Chu
Abstract
Federated Learning (FL) enables collaborative training while protecting the privacy of participant data. However, typical centralized FL structures are vulnerable to malicious client attacks. To mitigate such vulnerabilities, blockchain technology has been used to develop a decentralized FL framework. Yet, this approach leads to significant transmission and computation overhead. In order to improve robustness as well as efficiency, we propose Multi-Population Federated Learning (MPFL), which optimizes blockchain-based federated learning by incorporating operators of multi-population genetic algorithm (MPGA), such as selection, crossover, mutation and migration. Moreover, we introduce a committee consensus mechanism tailored to MPFL, where honest clients are elected to govern each population, thereby facilitating intraspecific and interspecific competition. Our experiments demonstrate that MPFL outperforms various aggregation algorithms in achieving superior robustness, and it reduces the costs of transmission and computation compared to the blockchain-based FL framework.
BibTeX
@inproceedings{icassp2025_mpfladecentralis,
title = {MPFL: A Decentralised Federated Learning Framework Based on Multi-Population Genetic Algorithm},
author = {Wenqi Ding and Yuanchao Liu and Zhongjie Wang and Zheng Chu},
booktitle = {ICASSP 2025},
year = {2025}
}