TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Geom-GCN: Geometric Graph Convolutional Networks

Geom-GCN: Geometric Graph Convolutional Networks

Hongbin Pei, Bingzhe Wei, Kevin Chen-Chuan Chang, Yu Lei, Bo Yang

2020-02-13ICLR 2020 1Node Classification on Non-Homophilic (Heterophilic) GraphsRepresentation LearningNode Classification
PaperPDFCodeCode(official)CodeCode

Abstract

Message-passing neural networks (MPNNs) have been successfully applied to representation learning on graphs in a variety of real-world applications. However, two fundamental weaknesses of MPNNs' aggregators limit their ability to represent graph-structured data: losing the structural information of nodes in neighborhoods and lacking the ability to capture long-range dependencies in disassortative graphs. Few studies have noticed the weaknesses from different perspectives. From the observations on classical neural network and network geometry, we propose a novel geometric aggregation scheme for graph neural networks to overcome the two weaknesses. The behind basic idea is the aggregation on a graph can benefit from a continuous space underlying the graph. The proposed aggregation scheme is permutation-invariant and consists of three modules, node embedding, structural neighborhood, and bi-level aggregation. We also present an implementation of the scheme in graph convolutional networks, termed Geom-GCN (Geometric Graph Convolutional Networks), to perform transductive learning on graphs. Experimental results show the proposed Geom-GCN achieved state-of-the-art performance on a wide range of open datasets of graphs. Code is available at https://github.com/graphdml-uiuc-jlu/geom-gcn.

Results

TaskDatasetMetricValueModel
Node ClassificationWisconsinAccuracy64.12Geom-GCN-P
Node ClassificationWisconsinAccuracy58.24Geom-GCN-I
Node ClassificationWisconsinAccuracy56.67Geom-GCN-S
Node ClassificationTexas (60%/20%/20% random splits)1:1 Accuracy67.57Geom-GCN*
Node ClassificationSquirrelAccuracy38.14Geom-GCN-P
Node ClassificationSquirrelAccuracy36.24Geom-GCN-S
Node ClassificationSquirrelAccuracy33.32Geom-GCN-I
Node ClassificationSquirrel (60%/20%/20% random splits)1:1 Accuracy38.14Geom-GCN*
Node ClassificationTexasAccuracy67.57Geom-GCN-P
Node ClassificationTexasAccuracy59.73Geom-GCN-S
Node ClassificationTexasAccuracy57.58Geom-GCN-I
Node ClassificationCornellAccuracy60.81Geom-GCN-P
Node ClassificationCornellAccuracy56.76Geom-GCN-I
Node ClassificationCornellAccuracy55.68Geom-GCN-S
Node ClassificationChameleon (60%/20%/20% random splits)1:1 Accuracy60.9Geom-GCN*
Node ClassificationChameleonAccuracy60.9Geom-GCN-P
Node ClassificationChameleonAccuracy60.31Geom-GCN-I
Node ClassificationChameleonAccuracy59.96Geom-GCN-S
Node ClassificationCornell (60%/20%/20% random splits)1:1 Accuracy60.81Geom-GCN*
Node ClassificationPubMed (60%/20%/20% random splits)1:1 Accuracy90.05Geom-GCN*
Node ClassificationFilm (60%/20%/20% random splits)1:1 Accuracy31.63Geom-GCN*
Node ClassificationWisconsin (60%/20%/20% random splits)1:1 Accuracy64.12Geom-GCN*
Node ClassificationCiteSeer (60%/20%/20% random splits)1:1 Accuracy77.99Geom-GCN*
Node ClassificationActorAccuracy31.63Geom-GCN-P
Node ClassificationActorAccuracy30.3Geom-GCN-S
Node ClassificationActorAccuracy29.09Geom-GCN-I
Node ClassificationCora (60%/20%/20% random splits)1:1 Accuracy85.27Geom-GCN*
Node ClassificationCornell (60%/20%/20% random splits)1:1 Accuracy60.81Geom-GCN*
Node ClassificationChameleon(60%/20%/20% random splits)1:1 Accuracy60.9Geom-GCN*
Node ClassificationTexas(60%/20%/20% random splits)1:1 Accuracy67.57Geom-GCN*
Node ClassificationWisconsin(60%/20%/20% random splits)1:1 Accuracy64.12Geom-GCN*

Related Papers

Touch in the Wild: Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper2025-07-20Spectral Bellman Method: Unifying Representation and Exploration in RL2025-07-17Boosting Team Modeling through Tempo-Relational Representation Learning2025-07-17Similarity-Guided Diffusion for Contrastive Sequential Recommendation2025-07-16Are encoders able to learn landmarkers for warm-starting of Hyperparameter Optimization?2025-07-16Language-Guided Contrastive Audio-Visual Masked Autoencoder with Automatically Generated Audio-Visual-Text Triplets from Videos2025-07-16A Mixed-Primitive-based Gaussian Splatting Method for Surface Reconstruction2025-07-15Dual Dimensions Geometric Representation Learning Based Document Dewarping2025-07-11