Contrastive Adaptive Propagation Graph Neural Networks for Efficient Graph Learning

TL;DR

This paper introduces CAPGNN, a novel graph neural network framework that adaptively combines local and high-order neighbor information using personalized PageRank, attention mechanisms, and contrastive learning, leading to improved graph learning performance.

Contribution

It proposes an end-to-end adaptive propagation scheme with a coefficient-attention model and contrastive learning, enhancing efficiency and effectiveness over existing GNN methods.

Findings

CAPGNN outperforms state-of-the-art baselines on benchmark datasets.

The coefficient-attention model effectively balances local and high-order neighbor influence.

Self-supervised contrastive learning improves training with unlabeled data.

Abstract

Graph Neural Networks (GNNs) have achieved great success in processing graph data by extracting and propagating structure-aware features. Existing GNN research designs various propagation schemes to guide the aggregation of neighbor information. Recently the field has advanced from local propagation schemes that focus on local neighbors towards extended propagation schemes that can directly deal with extended neighbors consisting of both local and high-order neighbors. Despite the impressive performance, existing approaches are still insufficient to build an efficient and learnable extended propagation scheme that can adaptively adjust the influence of local and high-order neighbors. This paper proposes an efficient yet effective end-to-end framework, namely Contrastive Adaptive Propagation Graph Neural Networks (CAPGNN), to address these issues by combining Personalized PageRank and attention techniques. CAPGNN models the learnable extended propagation scheme with a polynomial of a sparse local affinity matrix, where the polynomial relies on Personalized PageRank to provide superior initial coefficients. In order to adaptively adjust the influence of both local and high-order neighbors, a coefficient-attention model is introduced to learn to adjust the coefficients of the polynomial. In addition, we leverage self-supervised learning techniques and design a negative-free entropy-aware contrastive loss to explicitly take advantage of unlabeled data for training. We implement CAPGNN as two different versions named CAPGCN and CAPGAT, which use static and dynamic sparse local affinity matrices, respectively. Experiments on graph benchmark datasets suggest that CAPGNN can consistently outperform or match state-of-the-art baselines. The source code is publicly available at https://github.com/hujunxianligong/CAPGNN.