README.md

Publications in NeurIPS 2019

Understanding Graph Neural Networks

• Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology
• GNN Explainer: A Tool for Post-hoc Explanation of Graph Neural Networks
• On the equivalence between graph isomorphism testing and function approximation with GNNs
• Understanding Attention and Generalization in Graph Neural Networks
• Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
• Provably Powerful Graph Networks
• Universal Invariant and Equivariant Graph Neural Networks

Graph-based Semi-Supervised Learning

• A Flexible Generative Framework for Graph-based Semi-supervised Learning
• Graph-Based Semi-Supervised Learning with Non-ignorable Non-response
• Graph Agreement Models for Semi-Supervised Learning
• Certifiable Robustness to Graph Perturbations

Scalability

• Layer-Dependent Importance Sampling for Training Deep and Large Graph Convolutional Networks
• Diffusion Improves Graph Learning
• End to end learning and optimization on graphs

Hypergraphs

• HyperGCN: A New Method for Training Graph Convolutional Networks on Hypergraphs
• Powerset Convolutional Neural Networks

Hyperbolic Geometry

• Hyperbolic Graph Convolutional Neural Networks
• Hyperbolic Graph Neural Networks

Combinatorial Problems

• Approximation Ratios of Graph Neural Networks for Combinatorial Problems
• Exact Combinatorial Optimization with Graph Convolutional Neural Networks

Computer Vision

• Guided Similarity Separation for Image Retrieval
• Adaptive GNN for Image Analysis and Editing
• PasteGAN: A Semi-Parametric Method to Generate Image from Scene Graph
• Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations
• Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks
• Multi-label Co-regularization for Semi-supervised Facial Action Unit Recognition
• Recurrent Space-time Graph Neural Networks

Visual Commonsense

• Heterogeneous Graph Learning for Visual Commonsense Reasoning
• Connective Cognition Network for Directional Visual Commonsense Reasoning

Kernel Methods

• Rethinking Kernel Methods for Node Representation Learning on Graphs
• Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels
• Wasserstein Weisfeiler-Lehman Kernels on graphs

Spectral Formulations

• DFNets: Spectral CNNs for Graphs with Feedback-looped Filters
• Stability of Graph Scattering Transforms

Structured Prediction

• Graph Structured Prediction Energy Networks
• Retrosynthesis Prediction with Conditional Graph Logic Network

Programs

• Learning Transferable Graph Exploration
• Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks

Molecules

• N-Gram Graph, A Novel Molecule Representation
• Generative Models for Graph-Based Protein Design

Graph Generation

• Efficient Graph Generation with Graph Recurrent Attention Networks
• Conditional Structure Generation through Graph Variational Generative Adversarial Nets
• Graph Normalizing Flows
• Semi-Implicit Graph Variational Auto-Encoders

Miscellaneous

• Variational Graph Recurrent Neural Networks
• Graph Transformer Networks
• Embedding Symbolic Knowledge into Deep Networks
• Combining Generative and Discriminative Models for Hybrid Inference
• D-VAE: A Variational Autoencoder for Directed Acyclic Graphs
• Graph-based Discriminators: Sample Complexity and Expressiveness
• Scalable Deep Generative Relational Model with High-Order Node Dependence
• Neural Relational Inference with Fast Modular Meta-learning
• Hyper-Graph-Network Decoders for Block Codes