NetGAN: Generating Graphs via Random Walks

Abstract

We propose NetGAN - the first implicit generative model for graphs able to mimic real-world networks. We pose the problem of graph generation as learning the distribution of biased random walks over the input graph. The proposed model is based on a stochastic neural network that generates discrete output samples and is trained using the Wasserstein GAN objective. NetGAN is able to produce graphs that exhibit well-known network patterns without explicitly specifying them in the model definition. At the same time, our model exhibits strong generalization properties, as highlighted by its competitive link prediction performance, despite not being trained specifically for this task. Being the first approach to combine both of these desirable properties, NetGAN opens exciting avenues for further research.

arXiv commentICML 2018

References in corpus (18)

• Adam: A Method for Stochastic Optimization
• Emergence of scaling in random networks
• Categorical Reparameterization with Gumbel-Softmax
• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Improved Training of Wasserstein GANs
• NIPS 2016 Tutorial: Generative Adversarial Networks
• Variational Graph Auto-Encoders
• BEGAN: Boundary Equilibrium Generative Adversarial Networks
• Generating Multi-label Discrete Patient Records using Generative Adversarial Networks
• Adversarial Learning for Neural Dialogue Generation
• GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution
• GraphGAN: Graph Representation Learning with Generative Adversarial Nets
• Maximum-Likelihood Augmented Discrete Generative Adversarial Networks
• Boundary-Seeking Generative Adversarial Networks
• Recurrent Topic-Transition GAN for Visual Paragraph Generation
• Modeling Social Networks with Node Attributes using the Multiplicative Attribute Graph Model
• Can GAN Learn Topological Features of a Graph?

Cited by in corpus (56)

• Graph Neural Networks: A Review of Methods and Applications
• MolGAN: An implicit generative model for small molecular graphs
• A Gentle Introduction to Deep Learning for Graphs
• Adversarial Attack and Defense on Graph Data: A Survey
• Molecular graph generation with Graph Neural Networks
• Deep Learning on Graphs: A Survey
• Generative Adversarial Networks for Spatio-temporal Data: A Survey
• Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classification
• Generative Models for Automatic Chemical Design
• D-VAE: A Variational Autoencoder for Directed Acyclic Graphs
• Self-Contrastive Learning with Hard Negative Sampling for Self-supervised Point Cloud Learning
• Representation Learning for Natural Language Processing
• CorrGAN: Sampling Realistic Financial Correlation Matrices Using Generative Adversarial Networks
• Hypergraph Convolution and Hypergraph Attention
• Graph Representation Learning for Road Type Classification
• Graphite: Iterative Generative Modeling of Graphs
• Node Injection Attacks on Graphs via Reinforcement Learning
• Hierarchical Graph-to-Graph Translation for Molecules
• Hybrid Quantum-Classical Graph Convolutional Network
• TG-GAN: Continuous-time Temporal Graph Generation with Deep Generative Models
• Labeled Graph Generative Adversarial Networks
• Constructions in combinatorics via neural networks
• TIGGER: Scalable Generative Modelling for Temporal Interaction Graphs
• Generating Tertiary Protein Structures via an Interpretative Variational Autoencoder
• Deep Generation of Heterogeneous Networks
• Semantic Bottleneck Scene Generation
• Secure Deep Graph Generation with Link Differential Privacy
• Concept-Oriented Deep Learning: Generative Concept Representations
• Modeling Graphs with Vertex Replacement Grammars
• Non-Parametric Graph Learning for Bayesian Graph Neural Networks
• GraphTune: A Learning-based Graph Generative Model with Tunable Structural Features
• Towards Interpretable Graph Modeling with Vertex Replacement Grammars
• EvoNet: A Neural Network for Predicting the Evolution of Dynamic Graphs
• CCGG: A Deep Autoregressive Model for Class-Conditional Graph Generation
• Adversarial Graph Disentanglement
• MEGAN: A Generative Adversarial Network for Multi-View Network Embedding
• Node Copying: A Random Graph Model for Effective Graph Sampling
• GraphTER: Unsupervised Learning of Graph Transformation Equivariant Representations via Auto-Encoding Node-wise Transformations
• Learning Theory Can (Sometimes) Explain Generalisation in Graph Neural Networks
• The Infinity Mirror Test for Graph Models
• The Atlas for the Aspiring Network Scientist
• SHADOWCAST: Controllable Graph Generation
• Graph Sparsification with Generative Adversarial Network
• Deep Learning on Attributed Graphs: A Journey from Graphs to Their Embeddings and Back
• Partition and Code: learning how to compress graphs
• Network Generation with Differential Privacy
• Adversarially-learned Inference via an Ensemble of Discrete Undirected Graphical Models
• TGG: Transferable Graph Generation for Zero-shot and Few-shot Learning
• Cut-Based Graph Learning Networks to Discover Compositional Structure of Sequential Video Data
• Knowledge as Invariance -- History and Perspectives of Knowledge-augmented Machine Learning
• TD-GEN: Graph Generation With Tree Decomposition
• Cyclic Graph Attentive Match Encoder (CGAME): A Novel Neural Network For OD Estimation
• Attributed Graph Modeling with Vertex Replacement Grammars
• Towards quantitative methods to assess network generative models
• Social Graph Restoration via Random Walk Sampling
• Can NetGAN be improved on short random walks?