ICML 2021 accepted papers on Graph learning
49 of the 1,184 papers accepted at ICML 2021 (4.1%) are on Graph learning, compared with 3.9% at ICML 2020. They fall into 4 topics.
See ICML 2021 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 · 23 papers
- Equivariant message passing for the prediction of tensorial properties and molecular spectra (253 citations)
- E(n) Equivariant Graph Neural Networks (102 citations)
- On Explainability of Graph Neural Networks via Subgraph Explorations (90 citations)
- GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training (73 citations)
- Weisfeiler and Lehman Go Topological: Message Passing Simplicial Networks (45 citations)
- Directional Graph Networks (38 citations)
- A Unified Lottery Ticket Hypothesis for Graph Neural Networks (27 citations)
- Breaking the Limits of Message Passing Graph Neural Networks (26 citations)
- Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth (23 citations)
- Generative Causal Explanations for Graph Neural Networks (22 citations)
- Graph Neural Networks Inspired by Classical Iterative Algorithms (13 citations)
- GLSearch: Maximum Common Subgraph Detection via Learning to Search (11 citations)
and 11 more in the interactive map.
Graph algorithms and clustering · 14 papers
- ProGraML: A Graph-based Program Representation for Data Flow Analysis and Compiler Optimizations (26 citations)
- Cyclically Equivariant Neural Decoders for Cyclic Codes (8 citations)
- Flow-based Attribution in Graphical Models: A Recursive Shapley Approach (7 citations)
- Explainable Automated Graph Representation Learning with Hyperparameter Importance (5 citations)
- Deep Latent Graph Matching (4 citations)
- Order Matters: Probabilistic Modeling of Node Sequence for Graph Generation (4 citations)
- DAGs with No Curl: An Efficient DAG Structure Learning Approach (3 citations)
- DeepWalking Backwards: From Embeddings Back to Graphs (1 citation)
- Local Algorithms for Finding Densely Connected Clusters (1 citation)
- Stochastic Iterative Graph Matching (1 citation)
- Online Graph Dictionary Learning
- Spectral vertex sparsifiers and pair-wise spanners over distributed graphs
and 2 more in the interactive map.
Graph self-supervised learning · 11 papers
- Graph Contrastive Learning Automated (93 citations)
- GraphDF: A Discrete Flow Model for Molecular Graph Generation (30 citations)
- Self-supervised Graph-level Representation Learning with Local and Global Structure (28 citations)
- How Framelets Enhance Graph Neural Networks (9 citations)
- Size-Invariant Graph Representations for Graph Classification Extrapolations (9 citations)
- Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization (8 citations)
- Link Prediction with Persistent Homology: An Interactive View (6 citations)
- Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach (6 citations)
- Graph Mixture Density Networks (5 citations)
- Improving Molecular Graph Neural Network Explainability with Orthonormalization and Induced Sparsity (5 citations)
- Directed Graph Embeddings in Pseudo-Riemannian Manifolds (3 citations)
