Anisotropic Graph Convolutional Network for Semi-supervised Learning

TL;DR

This paper introduces an anisotropic graph convolutional network that employs nonlinear functions to improve semi-supervised node classification by preventing over-smoothing, inspired by anisotropic diffusion techniques.

Contribution

It proposes a novel anisotropic GCN framework that captures informative features and mitigates over-smoothing in semi-supervised learning tasks.

Findings

Achieves better or comparable accuracy on citation and image datasets.

Effectively prevents over-smoothing in graph convolutional networks.

Demonstrates the benefit of nonlinear, anisotropic diffusion in graph learning.

Abstract

Graph convolutional networks learn effective node embeddings that have proven to be useful in achieving high-accuracy prediction results in semi-supervised learning tasks, such as node classification. However, these networks suffer from the issue of over-smoothing and shrinking effect of the graph due in large part to the fact that they diffuse features across the edges of the graph using a linear Laplacian flow. This limitation is especially problematic for the task of node classification, where the goal is to predict the label associated with a graph node. To address this issue, we propose an anisotropic graph convolutional network for semi-supervised node classification by introducing a nonlinear function that captures informative features from nodes, while preventing oversmoothing. The proposed framework is largely motivated by the good performance of anisotropic diffusion in image and geometry processing, and learns nonlinear representations based on local graph structure and node features. The effectiveness of our approach is demonstrated on three citation networks and two image datasets, achieving better or comparable classification accuracy results compared to the standard baseline methods.