{"title":"GRE^2-MDCL:通过多维对比学习增强图形表示嵌入功能","authors":"Kaizhe Fan, Quanjun Li","doi":"arxiv-2409.07725","DOIUrl":null,"url":null,"abstract":"Graph representation learning has emerged as a powerful tool for preserving\ngraph topology when mapping nodes to vector representations, enabling various\ndownstream tasks such as node classification and community detection. However,\nmost current graph neural network models face the challenge of requiring\nextensive labeled data, which limits their practical applicability in\nreal-world scenarios where labeled data is scarce. To address this challenge,\nresearchers have explored Graph Contrastive Learning (GCL), which leverages\nenhanced graph data and contrastive learning techniques. While promising,\nexisting GCL methods often struggle with effectively capturing both local and\nglobal graph structures, and balancing the trade-off between nodelevel and\ngraph-level representations. In this work, we propose Graph Representation\nEmbedding Enhanced via Multidimensional Contrastive Learning (GRE2-MDCL). Our\nmodel introduces a novel triple network architecture with a multi-head\nattention GNN as the core. GRE2-MDCL first globally and locally augments the\ninput graph using SVD and LAGNN techniques. It then constructs a\nmultidimensional contrastive loss, incorporating cross-network, cross-view, and\nneighbor contrast, to optimize the model. Extensive experiments on benchmark\ndatasets Cora, Citeseer, and PubMed demonstrate that GRE2-MDCL achieves\nstate-of-the-art performance, with average accuracies of 82.5%, 72.5%, and\n81.6% respectively. Visualizations further show tighter intra-cluster\naggregation and clearer inter-cluster boundaries, highlighting the\neffectiveness of our framework in improving upon baseline GCL models.","PeriodicalId":501301,"journal":{"name":"arXiv - CS - Machine Learning","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"GRE^2-MDCL: Graph Representation Embedding Enhanced via Multidimensional Contrastive Learning\",\"authors\":\"Kaizhe Fan, Quanjun Li\",\"doi\":\"arxiv-2409.07725\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Graph representation learning has emerged as a powerful tool for preserving\\ngraph topology when mapping nodes to vector representations, enabling various\\ndownstream tasks such as node classification and community detection. However,\\nmost current graph neural network models face the challenge of requiring\\nextensive labeled data, which limits their practical applicability in\\nreal-world scenarios where labeled data is scarce. To address this challenge,\\nresearchers have explored Graph Contrastive Learning (GCL), which leverages\\nenhanced graph data and contrastive learning techniques. While promising,\\nexisting GCL methods often struggle with effectively capturing both local and\\nglobal graph structures, and balancing the trade-off between nodelevel and\\ngraph-level representations. In this work, we propose Graph Representation\\nEmbedding Enhanced via Multidimensional Contrastive Learning (GRE2-MDCL). Our\\nmodel introduces a novel triple network architecture with a multi-head\\nattention GNN as the core. GRE2-MDCL first globally and locally augments the\\ninput graph using SVD and LAGNN techniques. It then constructs a\\nmultidimensional contrastive loss, incorporating cross-network, cross-view, and\\nneighbor contrast, to optimize the model. Extensive experiments on benchmark\\ndatasets Cora, Citeseer, and PubMed demonstrate that GRE2-MDCL achieves\\nstate-of-the-art performance, with average accuracies of 82.5%, 72.5%, and\\n81.6% respectively. Visualizations further show tighter intra-cluster\\naggregation and clearer inter-cluster boundaries, highlighting the\\neffectiveness of our framework in improving upon baseline GCL models.\",\"PeriodicalId\":501301,\"journal\":{\"name\":\"arXiv - CS - Machine Learning\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Machine Learning\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07725\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Machine Learning","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07725","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
GRE^2-MDCL: Graph Representation Embedding Enhanced via Multidimensional Contrastive Learning
Graph representation learning has emerged as a powerful tool for preserving
graph topology when mapping nodes to vector representations, enabling various
downstream tasks such as node classification and community detection. However,
most current graph neural network models face the challenge of requiring
extensive labeled data, which limits their practical applicability in
real-world scenarios where labeled data is scarce. To address this challenge,
researchers have explored Graph Contrastive Learning (GCL), which leverages
enhanced graph data and contrastive learning techniques. While promising,
existing GCL methods often struggle with effectively capturing both local and
global graph structures, and balancing the trade-off between nodelevel and
graph-level representations. In this work, we propose Graph Representation
Embedding Enhanced via Multidimensional Contrastive Learning (GRE2-MDCL). Our
model introduces a novel triple network architecture with a multi-head
attention GNN as the core. GRE2-MDCL first globally and locally augments the
input graph using SVD and LAGNN techniques. It then constructs a
multidimensional contrastive loss, incorporating cross-network, cross-view, and
neighbor contrast, to optimize the model. Extensive experiments on benchmark
datasets Cora, Citeseer, and PubMed demonstrate that GRE2-MDCL achieves
state-of-the-art performance, with average accuracies of 82.5%, 72.5%, and
81.6% respectively. Visualizations further show tighter intra-cluster
aggregation and clearer inter-cluster boundaries, highlighting the
effectiveness of our framework in improving upon baseline GCL models.