Exploiting Node Content for Multiview Graph Convolutional Network and Adversarial Regularization

Abstract

Network representation learning (NRL) is crucial in the area of graph learning. Recently, graph autoencoders and its variants have gained much attention and popularity among various types of node embedding approaches. Most existing graph autoencoder-based methods aim to minimize the reconstruction errors of the input network while not explicitly considering the semantic relatedness between nodes. In this paper, we propose a novel network embedding method which models the consistency across different views of networks. More specifically, we create a second view from the input network which captures the relation between nodes based on node content and enforce the latent representations from the two views to be consistent by incorporating a multiview adversarial regularization module. The experimental studies on benchmark datasets prove the effectiveness of this method, and demonstrate that our method compares favorably with the state-of-the-art algorithms on challenging tasks such as link prediction and node clustering. We also evaluate our method on a real-world application, i.e., 30-day unplanned ICU readmission prediction, and achieve promising results compared with several baseline methods.

Graph representation learning based on deep generative gaussian mixture models

“…This regularizer is used to capture the information of the immediate neighbour of each node [33]. Lu et al [34] proposed MRGAE, a network embedding approach to model network consistency across different views. MRGAE generates a second view of the input network based on node content capturing the relationship between them.…”

“…Focusing on positive and negative encoder samplings, we present a novel perspective on the adversarial technique. Unlike the conventional adversarialbased approaches [6,33,34], which treated encoder-generated representations as negative samples, we address them as positive samples. Following our shift from negative samples to positive ones and changing the nature of the regularizer, we appraise the overall loss function.…”