ICML 2022 accepted papers on Graph learning
66 of the 1,233 papers accepted at ICML 2022 (5.4%) are on Graph learning, compared with 4.1% at ICML 2021. They fall into 4 topics.
See ICML 2022 in the interactive map
Submitting to ICML? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
GNN architectures · 28 papers
- Finding Global Homophily in Graph Neural Networks When Meeting Heterophily (37 citations)
- How Powerful are Spectral Graph Neural Networks (36 citations)
- Interpretable and Generalizable Graph Learning via Stochastic Attention Mechanism (25 citations)
- SE(3) Equivariant Graph Neural Networks with Complete Local Frames (21 citations)
- Local Augmentation for Graph Neural Networks (18 citations)
- Learning to Solve PDE-constrained Inverse Problems with Graph Networks (16 citations)
- A New Perspective on the Effects of Spectrum in Graph Neural Networks (8 citations)
- Graph-Coupled Oscillator Networks (8 citations)
- Equivariant Quantum Graph Circuits (7 citations)
- NAFS: A Simple yet Tough-to-beat Baseline for Graph Representation Learning (7 citations)
- GNNRank: Learning Global Rankings from Pairwise Comparisons via Directed Graph Neural Networks (6 citations)
- Going Deeper into Permutation-Sensitive Graph Neural Networks (6 citations)
and 16 more in the interactive map.
Graph self-supervised learning · 17 papers
- 3D Infomax improves GNNs for Molecular Property Prediction (108 citations)
- Rethinking Graph Neural Networks for Anomaly Detection (52 citations)
- ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning (47 citations)
- Structure-Aware Transformer for Graph Representation Learning (45 citations)
- G-Mixup: Graph Data Augmentation for Graph Classification (43 citations)
- Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations (25 citations)
- Let Invariant Rationale Discovery Inspire Graph Contrastive Learning (23 citations)
- Structural Entropy Guided Graph Hierarchical Pooling (20 citations)
- Learning from Counterfactual Links for Link Prediction (14 citations)
- TAM: Topology-Aware Margin Loss for Class-Imbalanced Node Classification (14 citations)
- Molecular Representation Learning via Heterogeneous Motif Graph Neural Networks (13 citations)
- Self-Supervised Representation Learning via Latent Graph Prediction (8 citations)
and 5 more in the interactive map.
Graph algorithms and clustering · 14 papers
- From block-Toeplitz matrices to differential equations on graphs: towards a general theory for scalable masked Transformers (7 citations)
- Boosting Graph Structure Learning with Dummy Nodes (4 citations)
- LeNSE: Learning To Navigate Subgraph Embeddings for Large-Scale Combinatorial Optimisation (3 citations)
- PACE: A Parallelizable Computation Encoder for Directed Acyclic Graphs (3 citations)
- A Hierarchical Transitive-Aligned Graph Kernel for Un-attributed Graphs
- Learning to Predict Graphs with Fused Gromov-Wasserstein Barycenters
- Sublinear-Time Clustering Oracle for Signed Graphs
- Weisfeiler-Lehman Meets Gromov-Wasserstein
- Entropic Causal Inference: Graph Identifiability
- Neuron Dependency Graphs: A Causal Abstraction of Neural Networks
- Practical Almost-Linear-Time Approximation Algorithms for Hybrid and Overlapping Graph Clustering
- Simultaneous Graph Signal Clustering and Graph Learning
and 2 more in the interactive map.
Knowledge and temporal graphs · 7 papers
- HousE: Knowledge Graph Embedding with Householder Parameterization (20 citations)
- Neural-Symbolic Models for Logical Queries on Knowledge Graphs (16 citations)
- Comprehensive Analysis of Negative Sampling in Knowledge Graph Representation Learning (9 citations)
- Cycle Representation Learning for Inductive Relation Prediction (5 citations)
- DSTAGNN: Dynamic Spatial-Temporal Aware Graph Neural Network for Traffic Flow Forecasting
- Understanding and Improving Knowledge Graph Embedding for Entity Alignment
- VarScene: A Deep Generative Model for Realistic Scene Graph Synthesis
