Papers › Heterogeneous Deep Graph Infomax

Heterogeneous Deep Graph Infomax

19 Nov 2019arXiv:1911.08538archive 2025-07-28

Yuxiang Ren, Bo Liu, Chao Huang, Peng Dai, Liefeng Bo, Jiawei Zhang

Graph representation learning is to learn universal node representations that preserve both node attributes and structural information. The derived node representations can be used to serve various downstream tasks, such as node classification and node clustering. When a graph is heterogeneous, the problem becomes more challenging than the homogeneous graph node learning problem. Inspired by the emerging information theoretic-based learning algorithm, in this paper we propose an unsupervised graph neural network Heterogeneous Deep Graph Infomax (HDGI) for heterogeneous graph representation learning. We use the meta-path structure to analyze the connections involving semantics in heterogeneous graphs and utilize graph convolution module and semantic-level attention mechanism to capture local representations. By maximizing local-global mutual information, HDGI effectively learns high-level node representations that can be utilized in downstream graph-related tasks. Experiment results show that HDGI remarkably outperforms state-of-the-art unsupervised graph representation learning methods on both classification and clustering tasks. By feeding the learned representations into a parametric model, such as logistic regression, we even achieve comparable performance in node classification tasks when comparing with state-of-the-art supervised end-to-end GNN models.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

ClassificationClusteringGeneral ClassificationGraph Neural NetworkGraph Representation LearningHeterogeneous Node ClassificationNode ClassificationNode ClusteringRepresentation Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-C Macro-F1 (20% training data) 90.94% #7 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-C Macro-F1 (80% training data) 91.53% #7 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-C Micro-F1 (20% training data) 91.75% #7 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-C Micro-F1 (80% training data) 92.26% #7 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-A Macro-F1 (20% training data) 89.88% #10 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-A Macro-F1 (80% training data) 91.06% #10 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-A Micro-F1 (20% training data) 90.62% #10 of 14 Archive leaderboard report
Heterogeneous Node Classification DBLP (PACT) 14k HDGI-A Micro-F1 (80% training data) 91.92% #10 of 14 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

ConvolutionGraph Neural Network

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections