Wasserstein Adversarially Regularized Graph Autoencoder

TL;DR

This paper presents WARGA, a novel graph autoencoder that uses Wasserstein regularization to improve node embeddings, demonstrating superior performance in link prediction and clustering tasks over existing methods.

Contribution

WARGA introduces Wasserstein adversarial regularization for graph autoencoders, enhancing latent space quality and outperforming models based on KL divergence and standard adversarial approaches.

Findings

WARGA outperforms state-of-the-art models in link prediction.

WARGA achieves better node clustering results.

The method effectively regularizes latent distributions using Wasserstein metric.

Abstract

This paper introduces Wasserstein Adversarially Regularized Graph Autoencoder (WARGA), an implicit generative algorithm that directly regularizes the latent distribution of node embedding to a target distribution via the Wasserstein metric. The proposed method has been validated in tasks of link prediction and node clustering on real-world graphs, in which WARGA generally outperforms state-of-the-art models based on Kullback-Leibler (KL) divergence and typical adversarial framework.