GraphGAN: Graph Representation Learning with Generative Adversarial Nets

Abstract

The goal of graph representation learning is to embed each vertex in a graph into a low-dimensional vector space. Existing graph representation learning methods can be classified into two categories: generative models that learn the underlying connectivity distribution in the graph, and discriminative models that predict the probability of edge existence between a pair of vertices. In this paper, we propose GraphGAN, an innovative graph representation learning framework unifying above two classes of methods, in which the generative model and discriminative model play a game-theoretical minimax game. Specifically, for a given vertex, the generative model tries to fit its underlying true connectivity distribution over all other vertices and produces "fake" samples to fool the discriminative model, while the discriminative model tries to detect whether the sampled vertex is from ground truth or generated by the generative model. With the competition between these two models, both of them can alternately and iteratively boost their performance. Moreover, when considering the implementation of generative model, we propose a novel graph softmax to overcome the limitations of traditional softmax function, which can be proven satisfying desirable properties of normalization, graph structure awareness, and computational efficiency. Through extensive experiments on real-world datasets, we demonstrate that GraphGAN achieves substantial gains in a variety of applications, including link prediction, node classification, and recommendation, over state-of-the-art baselines.

The 32nd AAAI Conference on Artificial Intelligence (AAAI 2018), 8 pages

References in corpus (4)

• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• struc2vec: Learning Node Representations from Structural Identity
• Adversarial Learning for Neural Dialogue Generation
• Aspect-augmented Adversarial Networks for Domain Adaptation

Cited by in corpus (27)

• MolGAN: An implicit generative model for small molecular graphs
• Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future
• NetGAN: Generating Graphs via Random Walks
• DKN: Deep Knowledge-Aware Network for News Recommendation
• Fast Gradient Attack on Network Embedding
• Network Representation Learning: A Survey
• Graphite: Iterative Generative Modeling of Graphs
• A survey on Adversarial Recommender Systems: from Attack/Defense strategies to Generative Adversarial Networks
• Bayesian graph convolutional neural networks via tempered MCMC
• Can Adversarial Network Attack be Defended?
• Enhancing Social Recommendation with Adversarial Graph Convolutional Networks
• Adversarial Classifier for Imbalanced Problems
• Network Embedding via Deep Prediction Model
• GraphAttacker: A General Multi-Task GraphAttack Framework
• Deep Hypergraph U-Net for Brain Graph Embedding and Classification
• Node Copying: A Random Graph Model for Effective Graph Sampling
• An End-to-End Neighborhood-based Interaction Model for Knowledge-enhanced Recommendation
• CARL: Content-Aware Representation Learning for Heterogeneous Networks
• A Structural Representation Learning for Multi-relational Networks
• Empirical effect of graph embeddings on fraud detection/ risk mitigation
• Adversarially-learned Inference via an Ensemble of Discrete Undirected Graphical Models
• Adversarial Training Methods for Network Embedding
• Network representation learning: A macro and micro view
• MGA: Momentum Gradient Attack on Network
• Learning Based Proximity Matrix Factorization for Node Embedding
• Multi-Hot Compact Network Embedding
• Multiple Graph Adversarial Learning