Graph Convolution: A High-Order and Adaptive Approach

TL;DR

This paper introduces HA-GCN, a versatile graph neural network framework with high-order and adaptive modules, improving performance on node classification, molecule property prediction, and molecule generation tasks.

Contribution

The paper proposes a novel high-order and adaptive graph convolutional framework, HA-GCN, with new modules for enhanced graph modeling capabilities.

Findings

Outperforms state-of-the-art on node classification

Achieves better molecule property prediction

Generates 32% more real molecules in generation tasks

Abstract

In this paper, we presented a novel convolutional neural network framework for graph modeling, with the introduction of two new modules specially designed for graph-structured data: the $k$-th order convolution operator and the adaptive filtering module. Importantly, our framework of High-order and Adaptive Graph Convolutional Network (HA-GCN) is a general-purposed architecture that fits various applications on both node and graph centrics, as well as graph generative models. We conducted extensive experiments on demonstrating the advantages of our framework. Particularly, our HA-GCN outperforms the state-of-the-art models on node classification and molecule property prediction tasks. It also generates 32% more real molecules on the molecule generation task, both of which will significantly benefit real-world applications such as material design and drug screening.