This repo contains a comprehensive compilation of graph and/or GNN papers that were accepted at the International Conference on Machine Learning 2024. Graph or Geometric machine learning possesses an indispensable role within the domain of machine learning research, providing invaluable insights, methodologies, and solutions to a diverse array of challenges and problems.
Short Overview: We've got around 250 papers focusing on Graphs and GNNs in ICML'24. The core themes of this year include equivariant GNNs, OODs, diffusions, heterophily, expressivity, and clustering. There's also a good amount of casual graph works, more than I expected. We have some very good physics-inspired research too. A lot of application papers are available, although I expected to see more in molecular, chemical GNNs and GFlowNets. Reinforcement learning also had a good boost this year. Have a look and throw me a review (and, a star ⭐, maybe!) Thanks!
View Topic list!
- GNN Theories
- Weisfeiler Leman
- Heterophily
- Hypergraph
- Expressivity
- Generalization
- Equivariant Graph Neural Networks
- Out-of-Distribution
- Diffusion
- Graph Matching
- Contrastive Learning
- Clustering
- Foundational Models
- Message Passing Neural Networks
- Graph Transformers
- Class Imbalance
- Optimal Transport
- Graph Generation
- Unsupervised Learning
- Meta-learning
- Disentanglement
- Others
- GNNs for PDE/ODE/Physics
- Graph and Large Language Models/Agents
- Knowledge Graph and Knowledge Graph Embeddings
- GNN Applications
- Spatial and/or Temporal GNNs
- Explainable AI
- Reinforcement Learning
- Graphs, Molecules and Biology
- GFlowNets
- Casual Discovery and Graphs
- Federated Learning, Privacy, Decentralization
- Scene Graphs
- Position Papers
- Others
- Weisfeiler-Leman at the margin: When more expressivity matters
- Aligning Transformers with Weisfeiler-Leman
- Weisfeiler Leman for Euclidean Equivariant Machine Learning
- Understanding Heterophily for Graph Neural Networks
- How Universal Polynomial Bases Enhance Spectral Graph Neural Networks: Heterophily, Over-smoothing, and Over-squashing
- Unraveling the Impact of Heterophilic Structures on Graph Positive-Unlabeled Learning
- Multi-Track Message Passing: Tackling Oversmoothing and Oversquashing in Graph Learning via Preventing Heterophily Mixing
- Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs
- Mitigating Oversmoothing Through Reverse Process of GNNs for Heterophilic Graphs
- Fast Algorithms for Hypergraph PageRank with Applications to Semi-Supervised Learning
- Hypergraph-enhanced Dual Semi-supervised Graph Classification
- Relational Learning in Pre-Trained Models: A Theory from Hypergraph Recovery Perspective
- Expressivity and Generalization: Fragment-Biases for Molecular GNNs
- On the Expressive Power of Spectral Invariant Graph Neural Networks
- The Expressive Power of Path-Based Graph Neural Networks
- On the Expressive Power of Spectral Invariant Graph Neural Networks
- An Empirical Study of Realized GNN Expressiveness
- Weisfeiler-Leman at the margin: When more expressivity matters
- Generalization Error of Graph Neural Networks in the Mean-field Regime
- On the Generalization of Equivariant Graph Neural Networks
- Enhancing Size Generalization in Graph Neural Networks through Disentangled Representation Learning
- What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional Encoding
- Semantically-correlated memories in a dense associative model
- Improved Stability and Generalization Guarantees of the Decentralized SGD Algorithm
- PAC-Bayesian Generalization Bounds for Knowledge Graph Representation Learning
- Beyond the Federation: Topology-aware Federated Learning for Generalization to Unseen Clients
- A Circuit Domain Generalization Framework for Efficient Logic Synthesis in Chip Design
- On the Generalization of Equivariant Graph Neural Networks
- Improving Equivariant Graph Neural Networks on Large Geometric Graphs via Virtual Nodes Learning
- Equivariant Graph Neural Operator for Modeling 3D Dynamics
- Generalist Equivariant Transformer Towards 3D Molecular Interaction Learning
- Topological Neural Networks go Persistent, Equivariant, and Continuous
- Interpreting Equivariant Representations
- Graph Automorphism Group Equivariant Neural Networks
- Weisfeiler Leman for Euclidean Equivariant Machine Learning
- Subequivariant Reinforcement Learning in 3D Multi-Entity Physical Environments
- Graph Structure Extrapolation for Out-of-Distribution Generalization
- When and How Does In-Distribution Label Help Out-of-Distribution Detection?
- Graph Out-of-Distribution Detection Goes Neighborhood Shaping
- Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design
- Disentangled Graph Self-supervised Learning for Out-of-Distribution Generalization
- Bounded and Uniform Energy-based Out-of-distribution Detection for Graphs
- Prometheus: Out-of-distribution Fluid Dynamics Modeling with Disentangled Graph ODE
- S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning
- Perfect Alignment May be Poisonous to Graph Contrastive Learning
- UniCorn: A Unified Contrastive Learning Approach for Multi-view Molecular Representation Learning
- Community-Invariant Graph Contrastive Learning
- Graph Adversarial Diffusion Convolution
- Editing Partially Observable Networks via Graph Diffusion Models
- Cluster-Aware Similarity Diffusion for Instance Retrieval
- Neurodegenerative Brain Network Classification via Adaptive Diffusion with Temporal Regularization
- Hyperbolic Geometric Latent Diffusion Model for Graph Generation
- Diffuse, Sample, Project: Plug-And-Play Controllable Graph Generation
- Graph Generation with Diffusion Mixture
- Learning Iterative Reasoning through Energy Diffusion
- Multi-View Clustering by Inter-cluster Connectivity Guided Reward
- Dynamic Spectral Clustering with Provable Approximation Guarantee
- EDISON: Enhanced Dictionary-Induced Tensorized Incomplete Multi-View Clustering with Gaussian Error Rank Minimization
- Pruned Pivot: Correlation Clustering Algorithm for Dynamic, Parallel, and Local Computation Models
- LSEnet: Lorentz Structural Entropy Neural Network for Deep Graph Clustering
- A Near-Linear Time Approximation Algorithm for Beyond-Worst-Case Graph Clustering
- Cluster-Aware Similarity Diffusion for Instance Retrieval
- Combinatorial Approximations for Cluster Deletion: Simpler, Faster, and Better
- Verifying message-passing neural networks via topology-based bounds tightening
- Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs
- PANDA: Expanded Width-Aware Message Passing Beyond Rewiring
- Pluvial Flood Emulation with Hydraulics-informed Message Passing
- On dimensionality of feature vectors in MPNNs
- What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional Encoding
- Triplet Interaction Improves Graph Transformers: Accurate Molecular Graph Learning with Triplet Graph Transformers
- Aligning Transformers with Weisfeiler-Leman
- Less is More: on the Over-Globalizing Problem in Graph Transformers
- Comparing Graph Transformers via Positional Encodings
- Subgraphormer: Unifying Subgraph GNNs and Graph Transformers via Graph Products
- Class-Imbalanced Graph Learning without Class Rebalancing
- Automated Loss function Search for Class-imbalanced Node Classification
- OT-CLIP: Understanding and Generalizing CLIP via Optimal Transport
- Optimal Transport for Structure Learning Under Missing Data
- Parameter Estimation in DAGs from Incomplete Data via Optimal Transport
- Generalized Sobolev Transport for Probability Measures on a Graph
- Scene Graph Generation Strategy with Co-occurrence Knowledge and Learnable Term Frequency
- Hyperbolic Geometric Latent Diffusion Model for Graph Generation
- Diffuse, Sample, Project: Plug-And-Play Controllable Graph Generation
- Graph Generation with Diffusion Mixture
- On the Role of Edge Dependency in Graph Generative Models
- Unsupervised Episode Generation for Graph Meta-learning
- Unsupervised Representation Learning of Brain Activity via Bridging Voxel Activity and Functional Connectivity
- Unsupervised Parameter-free Simplicial Representation Learning with Scattering Transforms
- Tackling Prevalent Conditions in Unsupervised Combinatorial Optimization: Cardinality, Minimum, Covering, and More
- Prometheus: Out-of-distribution Fluid Dynamics Modeling with Disentangled Graph ODE
- Disentangled Graph Self-supervised Learning for Out-of-Distribution Generalization
- Enhancing Size Generalization in Graph Neural Networks through Disentangled Representation Learning
- Disentangled Continual Graph Neural Architecture Search with Invariant Modular Supernet
- Enhancing Size Generalization in Graph Neural Networks through Disentangled Representation Learning
- GNNs Also Deserve Editing, and They Need It More Than Once
- Learning Divergence Fields for Shift-Robust Graph Representations
- Probabilistic Routing for Graph-Based Approximate Nearest Neighbor Search
- Efficient Contextual Bandits with Uninformed Feedback Graphs
- Efficient Contrastive Learning for Fast and Accurate Inference on Graphs
- Graph Geometry-Preserving Autoencoders
- Stereographic Spherical Sliced Wasserstein Distances
- Perfect Alignment May be Poisonous to Graph Contrastive Learning
- Sparse-IFT: Sparse Iso-FLOP Transformations for Maximizing Training Efficiency
- Prospector Heads: Generalized Feature Attribution for Large Models & Data
- Collective Certified Robustness against Graph Injection Attacks
- Graph Distillation with Eigenbasis Matching
- Graph Neural Networks Use Graphs When They Shouldn't
- Quantum Positional Encodings for Graph Neural Networks
- SLOG: An Inductive Spectral Graph Neural Network Beyond Polynomial Filter
- How Interpretable Are Interpretable Graph Neural Networks?
- Graph Neural Networks with a Distribution of Parametrized Graphs
- Cooperative Graph Neural Networks
- Generalization Error of Graph Neural Networks in the Mean-field Regime
- Networked Inequality: Preferential Attachment Bias in Graph Neural Network Link Prediction
- EvoluNet: Advancing Dynamic Non-IID Transfer Learning on Graphs
- How Graph Neural Networks Learn: Lessons from Training Dynamics
- Homomorphism Counts for Graph Neural Networks: All About That Basis
- HGCN2SP: Hierarchical Graph Convolutional Network for Two-Stage Stochastic Programming
- Faster Streaming and Scalable Algorithms for Finding Directed Dense Subgraphs in Large Graphs
- SiBBlInGS: Similarity-driven Building-Block Inference using Graphs across States
- Navigating Complexity: Toward Lossless Graph Condensation via Expanding Window Matching
- Recurrent Distance Filtering for Graph Representation Learning
- Graph External Attention Enhanced Transformer
- A Graph is Worth $K$ Words: Euclideanizing Graph using Pure Transformer
- Convergence Guarantees for the DeepWalk Embedding on Block Models
- Delaunay Graph: Addressing Over-Squashing and Over-Smoothing Using Delaunay Triangulation
- Simulation of Graph Algorithms with Looped Transformers
- Efficient PAC Learnability of Dynamical Systems Over Multilayer Networks
- Gaussian Processes on Cellular Complexes
- An Efficient Maximal Ancestral Graph Listing Algorithm
- Extending Test-Time Augmentation with Metamorphic Relations for Combinatorial Problems
- Graph Positional and Structural Encoder
- Translating Subgraphs to Nodes Makes Simple GNNs Strong and Efficient for Subgraph Representation Learning
- Exploring Correlations of Self-Supervised Tasks for Graphs
- Surprisingly Strong Performance Prediction with Neural Graph Features
- Bipartite Matching in Massive Graphs: A Tight Analysis of EDCS
- Uncertainty for Active Learning on Graphs
- Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders
- Graph Neural Stochastic Diffusion for Estimating Uncertainty in Node Classification
- Position Paper: Future Directions in the Theory of Graph Machine Learning
- Pairwise Alignment Improves Graph Domain Adaptation
- From Coarse to Fine: Enable Comprehensive Graph Self-supervised Learning with Multi-granular Semantic Ensemble
- Identifiability Matters: Revealing the Hidden Recoverable Condition in Unbiased Learning to Rank
- Graph Neural PDE Solvers with Conservation and Similarity-Equivariance
- Neural SPH: Improved Neural Modeling of Lagrangian Fluid Dynamics
- Locality-Sensitive Hashing-Based Efficient Point Transformer with Applications in High-Energy Physics
- PGODE: Towards High-quality System Dynamics Modeling
- Prometheus: Out-of-distribution Fluid Dynamics Modeling with Disentangled Graph ODE
- Equivariant Graph Neural Operator for Modeling 3D Dynamics
- HAMLET: Graph Transformer Neural Operator for Partial Differential Equations
- Towards General Algorithm Discovery for Combinatorial Optimization: Learning Symbolic Branching Policy from Bipartite Graph
- PDHG-Unrolled Learning-to-Optimize Method for Large-Scale Linear Programming
- Combinatorial Approximations for Cluster Deletion: Simpler, Faster, and Better
- Gaussian Plane-Wave Neural Operator for Electron Density Estimation
- Graph-enhanced Large Language Models in Asynchronous Plan Reasoning
- GPTSwarm: Language Agents as Optimizable Graphs
- Case-Based or Rule-Based: How Do Transformers Do the Math?
- SceneCraft: An LLM Agent for Synthesizing 3D Scenes as Blender Code
- LLaGA: Large Language and Graph Assistant
- MAGDi: Structured Distillation of Multi-Agent Interaction Graphs Improves Reasoning in Smaller Language Models
- CHEMREASONER: Heuristic Search over a Large Language Models Knowledge Space using Quantum-Chemical Feedback
- Latent Logic Tree Extraction for Event Sequence Explanation from LLMs
- Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization
- PAC-Bayesian Generalization Bounds for Knowledge Graph Representation Learning
- Knowledge Graphs Can be Learned with Just Intersection Features
- KnowFormer: Revisiting Transformers for Knowledge Graph Reasoning
- Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation
- Coarse-to-Fine Highlighting: Reducing Knowledge Hallucination in Large Language Models
- Knowledge-aware Reinforced Language Models for Protein Directed Evolution
- Unlocking the Power of Spatial and Temporal Information in Medical Multimodal Pre-training
- S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning
- Irregular Multivariate Time Series Forecasting: A Transformable Patching Graph Neural Networks Approach
- Neurodegenerative Brain Network Classification via Adaptive Diffusion with Temporal Regularization
- Temporal Spiking Neural Networks with Synaptic Delay for Graph Reasoning
- Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling
- Graph-based Time Series Clustering for End-to-End Hierarchical Forecasting
- Long Range Propagation on Continuous-Time Dynamic Graphs
- Biharmonic Distance of Graphs and its Higher-Order Variants: Theoretical Properties with Applications to Centrality and Clustering
- LaMAGIC: Language-Model-based Topology Generation for Analog Integrated Circuits
- SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signals
- Predicting and Interpreting Energy Barriers of Metallic Glasses with Graph Neural Networks
- CARTE: Pretraining and Transfer for Tabular Learning
- Neurodegenerative Brain Network Classification via Adaptive Diffusion with Temporal Regularization
- A Circuit Domain Generalization Framework for Efficient Logic Synthesis in Chip Design
- Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
- Locality-Sensitive Hashing-Based Efficient Point Transformer with Applications in High-Energy Physics
- Biharmonic Distance of Graphs and its Higher-Order Variants: Theoretical Properties with Applications to Centrality and Clustering
- Graph2Tac: Online Representation Learning of Formal Math Concepts
- The Merit of River Network Topology for Neural Flood Forecasting
- OT-CLIP: Understanding and Generalizing CLIP via Optimal Transport
- Bringing Motion Taxonomies to Continuous Domains via GPLVM on Hyperbolic manifolds
- Structure Your Data: Towards Semantic Graph Counterfactuals
- Temporal Spiking Neural Networks with Synaptic Delay for Graph Reasoning
- Federated Self-Explaining GNNs with Anti-shortcut Augmentations
- Explaining Graph Neural Networks via Structure-aware Interaction Index
- Generating In-Distribution Proxy Graphs for Explaining Graph Neural Networks
- EiG-Search: Generating Edge-Induced Subgraphs for GNN Explanation in Linear Time
- Graph Neural Network Explanations are Fragile
- Tackling Non-Stationarity in Reinforcement Learning via Causal-Origin Representation
- SiT: Symmetry-invariant Transformers for Generalisation in Reinforcement Learning
- HGAP: Boosting Permutation Invariant and Permutation Equivariant in Multi-Agent Reinforcement Learning via Graph Attention Network
- Subequivariant Reinforcement Learning in 3D Multi-Entity Physical Environments
- Breadth-First Exploration on Adaptive Grid for Reinforcement Learning
- Swallowing the Bitter Pill: Simplified Scalable Conformer Generation
- Representing Molecules as Random Walks Over Interpretable Grammars
- Expressivity and Generalization: Fragment-Biases for Molecular GNNs
- Generalist Equivariant Transformer Towards 3D Molecular Interaction Learning
- UniCorn: A Unified Contrastive Learning Approach for Multi-view Molecular Representation Learning
- Modelling Microbial Communities with Graph Neural Networks
- Structure-Aware E(3)-Invariant Molecular Conformer Aggregation Networks
- Projecting Molecules into Synthesizable Chemical Spaces
- Proteus: Exploring Protein Structure Generation for Enhanced Designability and Efficiency
- CHEMREASONER: Heuristic Search over a Large Language Models Knowledge Space using Quantum-Chemical Feedback
- Gaussian Plane-Wave Neural Operator for Electron Density Estimation
- Latent Logic Tree Extraction for Event Sequence Explanation from LLMs
- GFlowNet Training by Policy Gradients
- Embarrassingly Parallel GFlowNets
- Learning to Scale Logits for Temperature-Conditional GFlowNets
- Causal-IQA: Towards the Generalization of Image Quality Assessment Based on Causal Inference
- Optimal Transport for Structure Learning Under Missing Data
- Causal Representation Learning from Multiple Distributions: A General Setting
- Foundations of Testing for Finite-Sample Causal Discovery
- Optimal Kernel Choice for Score Function-based Causal Discovery
- Causal Discovery with Fewer Conditional Independence Tests
- How Transformers Learn Causal Structure with Gradient Descent
- From Geometry to Causality- Ricci Curvature and the Reliability of Causal Inference on Networks
- A Fixed-Point Approach for Causal Generative Modeling
- Neural Tangent Kernels Motivate Cross-Covariance Graphs in Neural Networks
- Scalable and Flexible Causal Discovery with an Efficient Test for Adjacency
- Discovering Mixtures of Structural Causal Models from Time Series Data
- Challenges and Considerations in the Evaluation of Bayesian Causal Discovery
- Causal Effect Identification in LiNGAM Models with Latent Confounders
- Adaptive Online Experimental Design for Causal Discovery
- Stable Differentiable Causal Discovery
- Federated Self-Explaining GNNs with Anti-shortcut Augmentations
- Beyond the Federation: Topology-aware Federated Learning for Generalization to Unseen Clients
- Effective Federated Graph Matching
- The Privacy Power of Correlated Noise in Decentralized Learning
- Privacy Attacks in Decentralized Learning
- Differentially Private Decentralized Learning with Random Walks
- SceneCraft: An LLM Agent for Synthesizing 3D Scenes as Blender Code
- Scene Graph Generation Strategy with Co-occurrence Knowledge and Learnable Term Frequency
- Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognition
- Position: Topological Deep Learning is the New Frontier for Relational Learning
- Position: Graph Foundation Models Are Already Here
- Position Paper: Future Directions in the Theory of Graph Machine Learning
- Position: Relational Deep Learning - Graph Representation Learning on Relational Databases
- Exploring the Enigma of Neural Dynamics Through A Scattering-Transform Mixer Landscape for Riemannian Manifold
- Open Ad Hoc Teamwork with Cooperative Game Theory
- Incorporating Information into Shapley Values: Reweighting via a Maximum Entropy Approach
- Predicting Lagrangian Multipliers for Mixed Integer Linear Programs
- MAGNOLIA: Matching Algorithms via GNNs for Online Value-to-go Approximation
- Dynamic Metric Embedding into lp Space
- Graph Mixup on Approximate GromovWasserstein Geodesics
- Differentiability and Optimization of Multiparameter Persistent Homology
- Graph-Triggered Rising Bandits
- On Interpolating Experts and Multi-Armed Bandits
- Multiplicative Weights Update, Area Convexity and Random Coordinate Descent for Densest Subgraph Problems
- Empowering Graph Invariance Learning with Deep Spurious Infomax
- Rethinking Independent Cross-Entropy Loss For Graph-Structured Data
- Predictive Coding beyond Correlations
- When is Transfer Learning Possible?
- Optimal Exact Recovery in Semi-Supervised Learning: A Study of Spectral Methods and Graph Convolutional Networks
- Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspective
- Graph As Point Set
- CKGConv: General Graph Convolution with Continuous Kernels
- REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updates
- VisionGraph: Leveraging Large Multimodal Models for Graph Theory Problems in Visual Context
- Finite Volume Features, Global Geometry Representations, and Residual Training for Deep Learning-based CFD Simulation
- Two Heads Are Better Than One: Boosting Graph Sparse Training via Semantic and Topological Awareness
- DUPLEX: Dual GAT for Complex Embedding of Directed Graphs
- MS-TIP: Imputation Aware Pedestrian Trajectory Prediction
- Learning Graph Representation via Graph Entropy Maximization
- QBMK: Quantum-based Matching Kernels for Un-attributed Graphs
- Mitigating Label Noise on Graphs via Topological Sample Selection
- Multi-View Stochastic Block Models
- Learning in Deep Factor Graphs with Gaussian Belief Propagation
- Individual Fairness in Graph Decomposition
- Graphon Mean Field Games with a Representative Player: Analysis and Learning Algorithm
- Sign Rank Limitations for Inner Product Graph Decoders
- Differentiable Mapper for Topological Optimization of Data Representation
- Incremental Topological Ordering and Cycle Detection with Predictions
Missing any paper? If any paper is absent from the list, please feel free to mail or open an issue or submit a pull request. I'll gladly add that! Also, If I mis-categorized, please knock!
Azmine Toushik Wasi
