V2EX  ›  英汉词典

Graph Neural Network

释义 Definition

图神经网络:一种用于处理图结构数据(由“节点”和“边”组成,如社交网络、分子结构、知识图谱)的深度学习模型。它通过在图上进行信息传递/聚合(message passing),学习节点、边或整张图的表示,用于分类、预测、检索等任务。(也常简称 GNN。)

发音 Pronunciation (IPA)

/ɡræf ˈnjʊrəl ˈnɛtˌwɜːrk/

例句 Examples

Graph neural networks can predict properties of molecules from their chemical structures.
图神经网络可以根据分子的化学结构预测其性质。

By aggregating information from neighboring nodes, a graph neural network learns representations that improve recommendation quality in large-scale social graphs.
通过聚合相邻节点的信息,图神经网络能够学习更有效的表示,从而提升大规模社交图中的推荐效果。

词源 Etymology

该术语由 graph(图) + neural network(神经网络) 组合而来:graph 源自希腊语 graphein(“书写、描绘”),在数学与计算机科学中引申为“由点和线构成的结构”;neural network 指受生物神经系统启发的模型结构。“Graph Neural Network”在学术界用于概括一类专门面向图数据的神经网络方法,近年因图卷积网络、消息传递框架等发展而广泛流行。

相关词 Related Words

文献与著作 Literary / Notable Works

  • Kipf & Welling, Semi-Supervised Classification with Graph Convolutional Networks (2017)
  • Hamilton, Ying & Leskovec, Inductive Representation Learning on Large Graphs (GraphSAGE) (2017)
  • Battaglia et al., Relational inductive biases, deep learning, and graph networks (2018)
  • William L. Hamilton, Graph Representation Learning (2020)
  • Wu et al., A Comprehensive Survey on Graph Neural Networks (2020)
关于   ·   帮助文档   ·   自助推广系统   ·   博客   ·   API   ·   FAQ   ·   Solana   ·   700 人在线   最高记录 6679   ·     Select Language
创意工作者们的社区
World is powered by solitude
VERSION: 3.9.8.5 · 14ms · UTC 21:39 · PVG 05:39 · LAX 13:39 · JFK 16:39
♥ Do have faith in what you're doing.