Scholarlink

Graphs provide a versatile representationRepresentation for a wide range of real-world data, including social networks, biologicalBiological molecules, text, and images. This chapter explores Graph Neural NetworksGraphgraph neural network (GNNs) as powerful tools for processing graph-structured data. It covers key tasksTask in graph analysis, such as graph-level property prediction, node classificationClassification, and edge prediction, highlighting their relevance across diverse applications. Beginning with foundational concepts like the graph Fourier transformGraphgraph Fourier transform, the chapter introduces spectralSpectral methods, including ChebNetChebNet and Graph Convolutional NetworksGraphgraph convolutional network (GCNs). It then discusses advanced architecturesArchitecture such as Graph Attention NetworksGraphgraph attention network (GATs) and general GNN frameworks. The chapter concludes with an examination of graph autoencodersGraphgraph autoencoder for unsupervisedUnsupervised learning on graphs, providing a comprehensive overview of modern techniques in graph-based deep Learningdeep learning learningDeepdeep learning.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Graph Neural Networks

  • 章节

  • 发表日期：

    2026年

查看全文
  • Benyamin Ghojogh,
  • Ali Ghodsi

摘要

Graphs provide a versatile representationRepresentation for a wide range of real-world data, including social networks, biologicalBiological molecules, text, and images. This chapter explores Graph Neural NetworksGraphgraph neural network (GNNs) as powerful tools for processing graph-structured data. It covers key tasksTask in graph analysis, such as graph-level property prediction, node classificationClassification, and edge prediction, highlighting their relevance across diverse applications. Beginning with foundational concepts like the graph Fourier transformGraphgraph Fourier transform, the chapter introduces spectralSpectral methods, including ChebNetChebNet and Graph Convolutional NetworksGraphgraph convolutional network (GCNs). It then discusses advanced architecturesArchitecture such as Graph Attention NetworksGraphgraph attention network (GATs) and general GNN frameworks. The chapter concludes with an examination of graph autoencodersGraphgraph autoencoder for unsupervisedUnsupervised learning on graphs, providing a comprehensive overview of modern techniques in graph-based deep Learningdeep learning learningDeepdeep learning.