HCL: Improving Graph Representation with Hierarchical Contrastive Learning
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.
Code (0)
등록된 구현이 없습니다.
Tasks
Contrastive LearningGraph ClassificationGraph Representation LearningNode ClassificationNode ClusteringRepresentation LearningMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Contrastive Hierarchical Discourse Graph for Scientific Document Summarization
The extended structural context has made scientific paper summarization a challenging task. This paper proposes CHANGES, a contrastive hierarchical graph neural network for extractive scientific paper summarization. CHAN…
Contrastive LearningDocument SummarizationGraph Neural NetworkScientific Document Summarization+1HGCL: Hierarchical Graph Contrastive Learning for User-Item Recommendation
Graph Contrastive Learning (GCL), which fuses graph neural networks with contrastive learning, has evolved as a pivotal tool in user-item recommendations. While promising, existing GCL methods often lack explicit modelin…
Contrastive LearningHierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning
Graph contrastive learning (GCL) aims to align the positive features while differentiating the negative features in the latent space by minimizing a pair-wise contrastive loss. As the embodiment of an outstanding discrim…
Contrastive LearningGraph Representation LearningRepresentation LearningTransfer LearningHiTeC: Hierarchical Contrastive Learning on Text-Attributed Hypergraph with Semantic-Aware Augmentation
Contrastive learning (CL) has become a dominant paradigm for self-supervised hypergraph learning, enabling effective training without costly labels. However, node entities in real-world hypergraphs are often associated w…
Self-Supervised LearningRepresentation LearningContrastive LearningHierarchical and Contrastive Representation Learning for Knowledge-aware Recommendation
Incorporating knowledge graph into recommendation is an effective way to alleviate data sparsity. Most existing knowledge-aware methods usually perform recursive embedding propagation by enumerating graph neighbors. Howe…
Contrastive LearningKnowledge-Aware RecommendationRepresentation Learning