Adversarial Graph Disentanglement

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zheng, Shuai, Zhu, Zhenfeng, Liu, Zhizhe, Cheng, Jian, Zhao, Yao
Format: Preprint
Publié: 2021
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917573292982272
author Zheng, Shuai
Zhu, Zhenfeng
Liu, Zhizhe
Cheng, Jian
Zhao, Yao
author_facet Zheng, Shuai
Zhu, Zhenfeng
Liu, Zhizhe
Cheng, Jian
Zhao, Yao
contents A real-world graph has a complex topological structure, which is often formed by the interaction of different latent factors. However, most existing methods lack consideration of the intrinsic differences in relations between nodes caused by factor entanglement. In this paper, we propose an \underline{\textbf{A}}dversarial \underline{\textbf{D}}isentangled \underline{\textbf{G}}raph \underline{\textbf{C}}onvolutional \underline{\textbf{N}}etwork (ADGCN) for disentangled graph representation learning. To begin with, we point out two aspects of graph disentanglement that need to be considered, i.e., micro-disentanglement and macro-disentanglement. For them, a component-specific aggregation approach is proposed to achieve micro-disentanglement by inferring latent components that cause the links between nodes. On the basis of micro-disentanglement, we further propose a macro-disentanglement adversarial regularizer to improve the separability among component distributions, thus restricting the interdependence among components. Additionally, to reveal the topological graph structure, a diversity-preserving node sampling approach is proposed, by which the graph structure can be progressively refined in a way of local structure awareness. The experimental results on various real-world graph data verify that our ADGCN obtains more favorable performance over currently available alternatives. The source codes of ADGCN are available at \textit{\url{https://github.com/SsGood/ADGCN}}.
format Preprint
id arxiv_https___arxiv_org_abs_2103_07295
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Adversarial Graph Disentanglement
Zheng, Shuai
Zhu, Zhenfeng
Liu, Zhizhe
Cheng, Jian
Zhao, Yao
Machine Learning
Artificial Intelligence
A real-world graph has a complex topological structure, which is often formed by the interaction of different latent factors. However, most existing methods lack consideration of the intrinsic differences in relations between nodes caused by factor entanglement. In this paper, we propose an \underline{\textbf{A}}dversarial \underline{\textbf{D}}isentangled \underline{\textbf{G}}raph \underline{\textbf{C}}onvolutional \underline{\textbf{N}}etwork (ADGCN) for disentangled graph representation learning. To begin with, we point out two aspects of graph disentanglement that need to be considered, i.e., micro-disentanglement and macro-disentanglement. For them, a component-specific aggregation approach is proposed to achieve micro-disentanglement by inferring latent components that cause the links between nodes. On the basis of micro-disentanglement, we further propose a macro-disentanglement adversarial regularizer to improve the separability among component distributions, thus restricting the interdependence among components. Additionally, to reveal the topological graph structure, a diversity-preserving node sampling approach is proposed, by which the graph structure can be progressively refined in a way of local structure awareness. The experimental results on various real-world graph data verify that our ADGCN obtains more favorable performance over currently available alternatives. The source codes of ADGCN are available at \textit{\url{https://github.com/SsGood/ADGCN}}.
title Adversarial Graph Disentanglement
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2103.07295