Learning Graph Embedding With Adversarial Training Methods

By Shirui Pan, Ruiqi Hu, Sai-Fu Fung, Guodong Long, Jing Jiang, Chengqi Zhang. IEEE Transactions on Cybernetics, 2019

347 citations recorded by the source corpus.

Graph embedding aims to transfer a graph into vectors to facilitate subsequent graph-analytics tasks like link prediction and graph clustering. Most approaches on graph embedding focus on preserving the graph structure or minimizing the reconstruction errors for graph data. They have mostly overlooked the embedding distribution of the latent codes, which unfortunately may lead to inferior representation in many cases. In this article, we present a novel adversarially regularized framework for graph embedding. By employing the graph convolutional network as an encoder, our framework embeds the topological information and node content into a vector representation, from which a graph decoder is further built to reconstruct the input graph. The adversarial training principle is applied to enforce our latent codes to match a prior Gaussian or uniform distribution. Based on this framework, we derive two variants of the adversarial models, the adversarially regularized graph autoencoder (ARGA) and its variational version, and adversarially regularized variational graph autoencoder (ARVGA), to learn the graph embedding effectively. We also exploit other potential variations of ARGA and ARVGA to get a deeper understanding of our designs. Experimental results that compared 12 algorithms for link prediction and 20 algorithms for graph clustering validate our solutions.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/4a751d520534771593d359069f99d7c4621a861a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistical and Nonlinear Physics

Citations in the corpus

A Comprehensive Survey on Graph Neural Networks (2019, methodology): “To further enforce the empirical distribution q ( Z | X , A ) approximate the prior distribution p ( Z ) , Adversarially Regularized Variational Graph Autoencoder (ARVGA) [62], [109] employs the training scheme of a generative adversarial networks (GAN) [110].”
Deep Graph Clustering via Dual Correlation Reduction (2021, methodology): “In contrast, by reducing the information correlation in a dual manner, DCRN can learn more meaningful representation to improve the clustering performance; 2) it can be observed that the GCN-based clustering methods GAE/VGAE (Kipf and Welling 2016b), ARGA (Pan et al. 2019) and DAEGC (Wang et al. 201”
A Survey on Deep Semi-Supervised Learning (2021, methodology): “ARVGA [79] VGAE zu zv Auv EqðZ jX;AÞ1⁄2log pðA jZÞ KL1⁄2qðZ jX;AÞkpðZÞ OðjVkEjÞ”
Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning (2021, background): “Some recent works try to improve GNNs in different directions, such as simplifying the computational complexity [37], [38], training with adversarial scheme [39], applying to large-scale graphs [40], [41], [42], and introducing novel operators [8], [43].”
Attributed Graph Clustering: A Deep Attentional Embedding Approach (2019, methodology): “, 2017], adversarially regularized method [Pan et al., 2019] or denoising autoencoder [Cao et al.”
Graph Learning: A Survey (2021, background): “Its successor, adversarially regularized variational graph autoencoder [154], adds a regularization process with an adversarial training approach to learn a more robust embedding.”