ICLR 2022 accepted papers on Graph learning
81 of the 1,095 papers accepted at ICLR 2022 (7.4%) are on Graph learning, compared with 4.4% at ICLR 2021. They fall into 4 topics.
See ICLR 2022 in the interactive map
Submitting to ICLR? 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 · 43 papers
- Is Homophily a Necessity for Graph Neural Networks? (55 citations)
- Graph-less Neural Networks: Teaching Old MLPs New Tricks Via Distillation (43 citations)
- How Attentive are Graph Attention Networks? (41 citations)
- Understanding over-squashing and bottlenecks on graphs via curvature (34 citations)
- From Stars to Subgraphs: Uplifting Any GNN with Local Structure Awareness (27 citations)
- Graph Neural Networks with Learnable Structural and Positional Representations (25 citations)
- Understanding and Improving Graph Injection Attack by Promoting Unnoticeability (24 citations)
- Graph Auto-Encoder via Neighborhood Wasserstein Reconstruction (23 citations)
- Equivariant and Stable Positional Encoding for More Powerful Graph Neural Networks (22 citations)
- Graph Condensation for Graph Neural Networks (19 citations)
- Equivariant Graph Mechanics Networks with Constraints (18 citations)
- GNN-LM: Language Modeling based on Global Contexts via GNN (18 citations)
and 31 more in the interactive map.
Graph self-supervised learning · 20 papers
- Pre-training Molecular Graph Representation with 3D Geometry (158 citations)
- Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis (59 citations)
- Discovering Invariant Rationales for Graph Neural Networks (58 citations)
- Chemical-Reaction-Aware Molecule Representation Learning (34 citations)
- Handling Distribution Shifts on Graphs: An Invariance Perspective (30 citations)
- Large-Scale Representation Learning on Graphs via Bootstrapping (30 citations)
- Automated Self-Supervised Learning for Graphs (21 citations)
- Data-Efficient Graph Grammar Learning for Molecular Generation (18 citations)
- Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations (18 citations)
- A Biologically Interpretable Graph Convolutional Network to Link Genetic Risk Pathways and Imaging Phenotypes of Disease (8 citations)
- Graph-Relational Domain Adaptation (8 citations)
- Simple GNN Regularisation for 3D Molecular Property Prediction and Beyond (8 citations)
and 8 more in the interactive map.
Graph algorithms and clustering · 10 papers
- What’s Wrong with Deep Learning in Tree Search for Combinatorial Optimization (10 citations)
- Graph Neural Network Guided Local Search for the Traveling Salesperson Problem (9 citations)
- Ada-NETS: Face Clustering via Adaptive Neighbour Discovery in the Structure Space (8 citations)
- Semi-relaxed Gromov-Wasserstein divergence and applications on graphs (3 citations)
- Topological Experience Replay (3 citations)
- Convergent Graph Solvers (1 citation)
- Graphon based Clustering and Testing of Networks: Algorithms and Theory (1 citation)
- On the Learning and Learnability of Quasimetrics (1 citation)
- End-to-End Learning of Probabilistic Hierarchies on Graphs
- Graph-based Nearest Neighbor Search in Hyperbolic Spaces
Knowledge and temporal graphs · 8 papers
- Learning to Remember Patterns: Pattern Matching Memory Networks for Traffic Forecasting (18 citations)
- NodePiece: Compositional and Parameter-Efficient Representations of Large Knowledge Graphs (13 citations)
- Predicting Physics in Mesh-reduced Space with Temporal Attention (11 citations)
- Relational Multi-Task Learning: Modeling Relations between Data and Tasks (7 citations)
- Query Embedding on Hyper-Relational Knowledge Graphs (3 citations)
- Explainable GNN-Based Models over Knowledge Graphs
- Inductive Relation Prediction Using Analogy Subgraph Embeddings
- Neural Relational Inference with Node-Specific Information
