Wireless Ad Hoc Federated Learning: A Fully Distributed Cooperative Machine Learning

TL;DR

This paper introduces WAFL, a fully distributed federated learning approach for mobile and sensor nodes that enables model training over opportunistic contacts without centralized servers, achieving high accuracy on Non-IID data.

Contribution

Proposes WAFL, a novel fully distributed federated learning framework for ad hoc networks, eliminating reliance on centralized servers and enabling effective model training over opportunistic contacts.

Findings

WAFL achieves 94.8-96.3% accuracy on benchmark datasets.

WAFL successfully trains models on highly-partitioned Non-IID data.

WAFL converges without centralized mechanisms in opportunistic networks.

Abstract

Privacy-sensitive data is stored in autonomous vehicles, smart devices, or sensor nodes that can move around with making opportunistic contact with each other. Federation among such nodes was mainly discussed in the context of federated learning with a centralized mechanism in many works. However, because of multi-vendor issues, those nodes do not want to rely on a specific server operated by a third party for this purpose. In this paper, we propose a wireless ad hoc federated learning (WAFL) -- a fully distributed cooperative machine learning organized by the nodes physically nearby. WAFL can develop generalized models from Non-IID datasets stored in distributed nodes locally by exchanging and aggregating them with each other over opportunistic node-to-node contacts. In our benchmark-based evaluation with various opportunistic networks, WAFL has achieved higher accuracy of 94.8-96.3% than the self-training case of 84.7%. All our evaluation results show that WAFL can train and converge the model parameters from highly-partitioned Non-IID datasets over opportunistic networks without any centralized mechanisms.