Adaptive Graph Convolutional Neural Networks

TL;DR

This paper introduces an adaptive graph convolutional neural network that learns task-specific graph structures during training, improving performance on diverse graph data.

Contribution

It proposes a flexible, data-driven approach to adaptively learn graph structures within CNNs, addressing variability in real-world graph data.

Findings

Improved convergence speed in training.

Enhanced predictive accuracy across datasets.

Effective learning of task-specific graph structures.

Abstract

Graph Convolutional Neural Networks (Graph CNNs) are generalizations of classical CNNs to handle graph data such as molecular data, point could and social networks. Current filters in graph CNNs are built for fixed and shared graph structure. However, for most real data, the graph structures varies in both size and connectivity. The paper proposes a generalized and flexible graph CNN taking data of arbitrary graph structure as input. In that way a task-driven adaptive graph is learned for each graph data while training. To efficiently learn the graph, a distance metric learning is proposed. Extensive experiments on nine graph-structured datasets have demonstrated the superior performance improvement on both convergence speed and predictive accuracy.