Node Classification On Cora With Public Split

评估指标

Accuracy

评测结果

各个模型在此基准测试上的表现结果

		Paper Title
OGC	86.9%	From Cluster Assumption to Graph Convolution: Graph-based Semi-Supervised Learning Revisited
GCN-TV	86.3%	Re-Think and Re-Design Graph Neural Networks in Spaces of Continuous Graph Diffusion Functionals
GCNII	85.5%	Simple and Deep Graph Convolutional Networks
GRAND	85.4 ± 0.4	Graph Random Neural Network for Semi-Supervised Learning on Graphs
CPF-ind-APPNP	85.3%	Extract the Knowledge of Graph Neural Networks and Go Beyond it: An Effective Knowledge Distillation Framework
GCN	85.1 ± 0.7	Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification
AIR-GCN	84.7%	GraphAIR: Graph Representation Learning with Neighborhood Aggregation and Interaction
H-GCN	84.5%	Hierarchical Graph Convolutional Networks for Semi-supervised Node Classification
DAGNN (Ours)	84.4 ± 0.5	Towards Deeper Graph Neural Networks
G-APPNP	84.31%	Pre-train and Learn: Preserve Global Information for Graph Neural Networks
SuperGAT MX	84.3%	How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision
DSGCN	84.2%	Bridging the Gap Between Spectral and Spatial Domains in Graph Neural Networks
LDS-GNN	84.1%	Learning Discrete Structures for Graph Neural Networks
GraphMix	83.94 ± 0.57	GraphMix: Improved Training of GNNs for Semi-Supervised Learning
GraphMix (GCN)	83.94 ± 0.57	GraphMix: Improved Training of GNNs for Semi-Supervised Learning
GGCM	83.6%	From Cluster Assumption to Graph Convolution: Graph-based Semi-Supervised Learning Revisited
GCN+GAugO	83.6 ± 0.5%	Data Augmentation for Graph Neural Networks
Snowball (linear)	83.26%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
GAT+PGN	83.26 ± 0.69%	The Split Matters: Flat Minima Methods for Improving the Performance of GNNs
Snowball (tanh)	83.19%	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks

0 of 36 row(s) selected.