Optimization of Graph Neural Networks with Natural Gradient Descent | ✓ Link | 90.16% ± 0.59% | | | | | SSP | 2020-08-21 |
SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels | ✓ Link | 89.48% ± 0.31% | | | | | SplineCNN | 2017-11-24 |
Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification? | | 89.36% ± 1.26% | | | | | ACMII-Snowball-3 | 2021-09-12 |
Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification? | | 88.95% ± 1.04% | | | | | ACMII-GCN | 2021-09-12 |
Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification? | | 88.83% ± 1.49% | | | | | ACM-Snowball-2 | 2021-09-12 |
Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role-based Similarity | ✓ Link | 88.74±0.6% | | | | | UGT | 2023-08-18 |
The Split Matters: Flat Minima Methods for Improving the Performance of GNNs | ✓ Link | 88.66 ± 1.38% | | | | | GAT + SWA | 2023-06-15 |
Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification? | | 88.62% ± 1.22% | | | | | ACM-GCN | 2021-09-12 |
Unifying Graph Convolutional Neural Networks and Label Propagation | ✓ Link | 88.5% ± 1.5% | | | | | GCN-LPA | 2020-02-17 |
Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages | ✓ Link | 88.24% | | | | | LinkDist | 2021-06-16 |
CN-Motifs Perceptive Graph Neural Networks | | 88.20±1.22% | | | | | CNMPGNN | 2021-11-15 |
Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages | ✓ Link | 87.89% | | | | | CoLinkDist | 2021-06-16 |
Inferring from References with Differences for Semi-Supervised Node Classification on Graphs | ✓ Link | 87.78% | | | | | 3ference | 2022-04-11 |
Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages | ✓ Link | 87.58% | | | | | LinkDistMLP | 2021-06-16 |
Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages | ✓ Link | 87.54% | | | | | CoLinkDistMLP | 2021-06-16 |
Adaptive Sampling Towards Fast Graph Representation Learning | ✓ Link | 87.44% ± 0.0034% | | | | | AS-GCN | 2018-09-14 |
Mitigating Degree Biases in Message Passing Mechanism by Utilizing Community Structures | ✓ Link | 87.10±1.53 | | | | | CGT | 2023-12-28 |
NodeNet: A Graph Regularised Neural Network for Node Classification | | 86.80% | | | | | NodeNet | 2020-06-16 |
Cleora: A Simple, Strong and Scalable Graph Embedding Scheme | ✓ Link | 86.80% | | | | | Cleora | 2021-02-03 |
DFNets: Spectral CNNs for Graphs with Feedback-Looped Filters | ✓ Link | 86% ± 0.4% | | | | | DFNet-ATT | 2019-10-24 |
Multi-Mask Aggregators for Graph Neural Networks | ✓ Link | 85.80% | | | | | MMA | 2022-11-24 |
GResNet: Graph Residual Network for Reviving Deep GNNs from Suspended Animation | ✓ Link | 85.5% | | | | | GResNet(GAT) | 2019-09-12 |
AdaGCN: Adaboosting Graph Convolutional Networks into Deep Models | ✓ Link | 85.46% ± 0.25% | | | | | AdaGCN | 2019-08-14 |
Tree Decomposed Graph Neural Network | ✓ Link | 85.35% ± 0.49% | | | | | TDGNN | 2021-08-25 |
Predict then Propagate: Graph Neural Networks meet Personalized PageRank | ✓ Link | 85.29% ± 0.25% | | YES | | | PPNP | 2018-10-14 |
Get Rid of Suspended Animation Problem: Deep Diffusive Neural Network on Graph Semi-Supervised Classification | ✓ Link | 85.1% | | | | | DifNet | 2020-01-22 |
Predict then Propagate: Graph Neural Networks meet Personalized PageRank | ✓ Link | 85.09% ± 0.25% | | YES | | | APPNP | 2018-10-14 |
Robust Graph Data Learning via Latent Graph Convolutional Representation | | 84.8% | | | | | GOCN | 2019-04-26 |
Graph U-Nets | ✓ Link | 84.4% ± 0.6% | | | | | Graph U-Nets | 2019-05-11 |
GResNet: Graph Residual Network for Reviving Deep GNNs from Suspended Animation | ✓ Link | 84.3% | | | | | GResNet(GCN) | 2019-09-12 |
Graph-Bert: Only Attention is Needed for Learning Graph Representations | ✓ Link | 84.3% | | | | | Graph-Bert | 2020-01-15 |
GraphNAS: Graph Neural Architecture Search with Reinforcement Learning | ✓ Link | 84.2% ± 1.0% | | | | | GraphNAS | 2019-04-22 |
Learning Discrete Structures for Graph Neural Networks | ✓ Link | 84.08 ± 0.4% | | | | | LDS-GNN | 2019-03-28 |
Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks | ✓ Link | 83.98% ± 0.52% | | | | | GAT (DGL) | 2019-09-03 |
GResNet: Graph Residual Network for Reviving Deep GNNs from Suspended Animation | ✓ Link | 83.9% | | | | | GResNet(LoopyNet) | 2019-09-12 |
DiffWire: Inductive Graph Rewiring via the Lovász Bound | ✓ Link | 83.66% | | | | | CT-Layer (PE) | 2022-06-15 |
Auto-GNN: Neural Architecture Search of Graph Neural Networks | | 83.6% ± 0.3% | | | | | AGNN-w/o share | 2019-09-07 |
Structure fusion based on graph convolutional networks for semi-supervised classification | | 83.5% | | | | | SPF-GCN | 2019-07-02 |
Every Node Counts: Self-Ensembling Graph Convolutional Networks for Semi-Supervised Learning | ✓ Link | 83.5% ± 0.4% | | | | | SEGCN | 2018-09-26 |
Large-Scale Learnable Graph Convolutional Networks | ✓ Link | 83.3% | | | | | LGCN sub | 2018-08-12 |
Structure fusion based on graph convolutional networks for semi-supervised classification | | 83.3% | | | | | SF-GCN | 2019-07-02 |
Deep Graph Contrastive Representation Learning | ✓ Link | 83.3% ± 0.4% | | | | | GRACE | 2020-06-07 |
hpGAT: High-order Proximity Informed Graph Attention Network | | 83.1% | | | | | hpGAT | 2019-08-28 |
Graph Attention Networks | ✓ Link | 83.0% ± 0.7% | fixed 20 per node | YES | | | GAT | 2017-10-30 |
N-GCN: Multi-scale Graph Convolution for Semi-supervised Node Classification | ✓ Link | 83.0% | | | | | N-GCN | 2018-02-24 |
Certifiable Robustness and Robust Training for Graph Convolutional Networks | ✓ Link | 83% | | | | | GNN RH-U | 2019-06-28 |
A Flexible Generative Framework for Graph-based Semi-supervised Learning | ✓ Link | 82.9% | 20 per node with early stopping set | YES | | | G3NN | 2019-05-26 |
FIT-GNN: Faster Inference Time for GNNs Using Coarsening | ✓ Link | 82.9% | | | | 0.0019 | FIT-GNN | 2024-10-19 |
Understanding over-squashing and bottlenecks on graphs via curvature | ✓ Link | 82.76±0.23% | | | | | SDRF | 2021-11-29 |
GResNet: Graph Residual Network for Reviving Deep GNNs from Suspended Animation | ✓ Link | 82.6% | | | | | LoopyNet | 2019-09-12 |
Graph Adversarial Training: Dynamically Regularizing Based on Graph Structure | ✓ Link | 82.6% | | | | | GraphVAT | 2019-02-20 |
Deep Graph Infomax | ✓ Link | 82.3 ± 0.6% | | | | | DGI | 2018-09-27 |
Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View | | 82.3% | | | | | GCN + AdaGraph (AG) | 2019-09-07 |
Fast Graph Representation Learning with PyTorch Geometric | ✓ Link | 82.2% ± 1.5% | | | | | APPNP | 2019-03-06 |
SNoRe: Scalable Unsupervised Learning of Symbolic Node Representations | ✓ Link | 82.2% | | | | | SNoRe | 2020-09-08 |
Graph Star Net for Generalized Multi-Task Learning | ✓ Link | 82.1% | | | | | GraphStar | 2019-06-21 |
Graphite: Iterative Generative Modeling of Graphs | ✓ Link | 82.1% ± 0.06% | | | | | Graphite | 2018-03-28 |
MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing | ✓ Link | 81.9% | 20 per node | YES | | | MixHop | 2019-04-30 |
Graph Convolutional Neural Networks via Scattering | ✓ Link | 81.9% | | | | | GraphScattering | 2018-03-31 |
Graph Wavelet Neural Network | ✓ Link | 81.6% | | | | | GWNN | 2019-04-12 |
Semi-Supervised Classification with Graph Convolutional Networks | ✓ Link | 81.5% | | | | | GCN | 2016-09-09 |
Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering | ✓ Link | 81.2% | | | | | ChebNet | 2016-06-30 |
Mutual Teaching for Graph Convolutional Networks | ✓ Link | 80.9% | | | | | MT-GCN | 2020-09-02 |
Graph-less Neural Networks: Teaching Old MLPs New Tricks via Distillation | ✓ Link | 80.54± 1.35% | | | | | GLNN | 2021-10-17 |
A Capsule Network-based Model for Learning Node Embeddings | ✓ Link | 80.53% | | | | | Caps2NE | 2019-11-12 |
Hyperbolic Graph Convolutional Neural Networks | ✓ Link | 79.9% | | | | | HGCN | 2019-10-28 |
Multi-Task Graph Autoencoders | ✓ Link | 79.00% | | YES | | | MTGAE | 2018-11-07 |
Learning to Make Predictions on Graphs with Autoencoders | ✓ Link | 78.30% | | | | | alpha-LoNGAE | 2018-02-23 |
Deeper-GXX: Deepening Arbitrary GNNs | | 76.99±1.13% | | | | | TGCL+ResNet | 2021-10-26 |
Revisiting Semi-Supervised Learning with Graph Embeddings | ✓ Link | 75.7% | | | | | Planetoid* | 2016-03-29 |
DiffWire: Inductive Graph Rewiring via the Lovász Bound | ✓ Link | 67.96% | | | | | CT-Layer | 2022-06-15 |
Watch Your Step: Learning Node Embeddings via Graph Attention | ✓ Link | 67.9% | | | | | AttentionWalk | 2017-10-26 |
Strong Transitivity Relations and Graph Neural Networks | ✓ Link | | | | 85.1 | | TransGNN | 2024-01-01 |