Wide and Deep Graph Neural Networks with Distributed Online Learning

TL;DR

This paper introduces WD-GNN, a novel architecture combining linear filters and GNNs, enabling efficient distributed online learning and adaptation to dynamic graph topologies, with proven convergence and practical validation on robot swarm control.

Contribution

The paper proposes WD-GNN, a new architecture that facilitates distributed online learning for GNNs, addressing topology changes and providing convergence guarantees.

Findings

Effective online retraining of the wide component improves GNN adaptability.

Decentralized online learning outperforms centralized methods in dynamic graph scenarios.

Experimental results on robot swarm control validate the approach's practicality.

Abstract

Graph neural networks (GNNs) learn representations from network data with naturally distributed architectures, rendering them well-suited candidates for decentralized learning. Oftentimes, this decentralized graph support changes with time due to link failures or topology variations. These changes create a mismatch between the graphs on which GNNs were trained and the ones on which they are tested. Online learning can be used to retrain GNNs at testing time, overcoming this issue. However, most online algorithms are centralized and work on convex problems (which GNNs rarely lead to). This paper proposes the Wide and Deep GNN (WD-GNN), a novel architecture that can be easily updated with distributed online learning mechanisms. The WD-GNN comprises two components: the wide part is a bank of linear graph filters and the deep part is a GNN. At training time, the joint architecture learns a nonlinear representation from data. At testing time, the deep part (nonlinear) is left unchanged, while the wide part is retrained online, leading to a convex problem. We derive convergence guarantees for this online retraining procedure and further propose a decentralized alternative. Experiments on the robot swarm control for flocking corroborate theory and show potential of the proposed architecture for distributed online learning.