HCL: Improving Graph Representation with Hierarchical Contrastive Learning
This work addresses the limitation of fixed-scale contrastive learning in underestimating local or global information for graph representation learning, offering an incremental improvement.
The authors tackled the problem of graph representation learning by proposing a Hierarchical Contrastive Learning (HCL) framework to capture hierarchical and richer representations, achieving competitive performance on 12 datasets for tasks like node classification, node clustering, and graph classification.
Contrastive learning has emerged as a powerful tool for graph representation learning. However, most contrastive learning methods learn features of graphs with fixed coarse-grained scale, which might underestimate either local or global information. To capture more hierarchical and richer representation, we propose a novel Hierarchical Contrastive Learning (HCL) framework that explicitly learns graph representation in a hierarchical manner. Specifically, HCL includes two key components: a novel adaptive Learning to Pool (L2Pool) method to construct more reasonable multi-scale graph topology for more comprehensive contrastive objective, a novel multi-channel pseudo-siamese network to further enable more expressive learning of mutual information within each scale. Comprehensive experimental results show HCL achieves competitive performance on 12 datasets involving node classification, node clustering and graph classification. In addition, the visualization of learned representation reveals that HCL successfully captures meaningful characteristics of graphs.