Graph neural networks (GNNs) are powerful tools for developing scalable, decentralized artificial intelligence in large-scale networked systems, such as wireless networks, power grids, and transportation networks. Currently, GNNs in networked systems mostly follow a paradigm of `centralized training, distributed execution', which limits their adaptability and slows down their development cycles. In this work, we fill this gap for the first time by developing a communication-efficient, fully distributed online training approach for GNNs applied to large networked systems. For a mini-batch with B samples, our approach of training an L-layer GNN only adds L rounds of message passing to the LB rounds required by GNN inference, with doubled message sizes. Through numerical experiments in graph-based node regression, power allocation, and link scheduling in wireless networks, we demonstrate the effectiveness of our approach in training GNNs under supervised, unsupervised, and reinforcement learning paradigms.
@article{arxiv.2412.06105,
title = {Fully Distributed Online Training of Graph Neural Networks in Networked Systems},
author = {Rostyslav Olshevskyi and Zhongyuan Zhao and Kevin Chan and Gunjan Verma and Ananthram Swami and Santiago Segarra},
journal= {arXiv preprint arXiv:2412.06105},
year = {2024}
}