Siamese Attribute-missing Graph Auto-encoder

TL;DR

This paper introduces SAGA, a novel graph auto-encoder that enhances attribute and structure embedding interaction for attribute-missing graphs, improving data completion through a siamese network and structural constraints.

Contribution

The paper proposes a siamese network-based auto-encoder with structural constraints to better integrate attribute and structure information in attribute-missing graphs.

Findings

Outperforms state-of-the-art methods on six benchmark datasets.

Effectively captures high-order discriminative features.

Improves latent feature quality for data completion.

Abstract

Graph representation learning (GRL) on attribute-missing graphs, which is a common yet challenging problem, has recently attracted considerable attention. We observe that existing literature: 1) isolates the learning of attribute and structure embedding thus fails to take full advantages of the two types of information; 2) imposes too strict distribution assumption on the latent space variables, leading to less discriminative feature representations. In this paper, based on the idea of introducing intimate information interaction between the two information sources, we propose our Siamese Attribute-missing Graph Auto-encoder (SAGA). Specifically, three strategies have been conducted. First, we entangle the attribute embedding and structure embedding by introducing a siamese network structure to share the parameters learned by both processes, which allows the network training to benefit from more abundant and diverse information. Second, we introduce a K-nearest neighbor (KNN) and structural constraint enhanced learning mechanism to improve the quality of latent features of the missing attributes by filtering unreliable connections. Third, we manually mask the connections on multiple adjacent matrices and force the structural information embedding sub-network to recover the true adjacent matrix, thus enforcing the resulting network to be able to selectively exploit more high-order discriminative features for data completion. Extensive experiments on six benchmark datasets demonstrate the superiority of our SAGA against the state-of-the-art methods.