标签独立信息压缩的异质图表示  

Label-independent Information Compression for Heterogeneous Graph Representation

作　　者：马剑 王怡菲 孟丽[1] 何云飞 杨飞[1] MA Jian;WANG Yifei;MENG Li;HE Yunfei;YANG Fei(School of Biomedical Engineering,Anhui Medical University,Hefei 230012,China)

机构地区：[1]安徽医科大学生物医学工程学院,安徽合肥230012

出　　处：《计算机与现代化》2025年第4期36-41,共6页Computer and Modernization

基　　金：国家自然科学基金资助项目(62306011);安徽省自然科学基金资助项目(2108085MH303);安徽省高校自然科学研究项目重点项目(2023AH050631);安徽医科大学研究生科研与实践创新项目(YJS20230147)。

摘　　要：现有异质图(Heterogeneous Graph,HG)表示方法大都基于强大的图神经网络,聚合元路径内及元路径之间的语义信息来嵌入节点。然而,现有方法忽视了HG中节点的异质性,导致邻居节点中的无关信息沿着复杂结构扩散到高阶节点,扰动HG表示。为克服该问题,本文提出一种标签独立信息压缩的异质图表示方法LICHGR(Label-independent Information Compression for Heterogeneous Graph Representation)。LICHGR的核心思想是在信息瓶颈的指导下,利用希尔伯特-斯密特独立性准则限制异质图中标签独立信息的传播而尽可能保留标签依赖的信息。具体地,LICHGR通过在输入特征、元路径内隐藏特征、真实标签之间构造多方面的标签独立信息压缩限制,抽取丰富的标签依赖的信息,从而提高异质图表示质量。在3个公开的数据集上设计的多个实验充分验证了LICHGR的有效性。The existing methods for heterogeneous graph(HG)representation are mostly based on powerful graph neural networks,which aggregate semantic information within and between meta-paths to embed nodes.However,these existing approaches overlook the heterogeneity of nodes in HG,causing irrelevant information from neighboring nodes to spread along graph structures to higher-order nodes,disturbing the HG representation.To overcome this problem,this paper proposes a heterogeneous graph representation method called Label-Independent Compression for Heterogeneous Graph(LICHGR).The core idea of LICHGR is,under the guidance of the Information Bottleneck,to utilize the Hilbert-Schmidt Independence Criterion to restrict the propagation of label-independent information in heterogeneous graph while preserving label-dependent information as much as possible.Specifically,LICHGR constructs multi-faceted label-independent compression constraints among input features,hidden features within meta-paths,and true labels,extracting rich label-dependent information to enhance the quality of heterogeneous graph representation.Multiple experiments designed on three public datasets validate the effectiveness of LICHGR.

关 键 词：图神经网络 异质图表示 信息瓶颈 希尔伯特-斯密特独立性准则 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]