Graph autoencoders (AE) and variational autoencoders (VAE) recently emerged as powerful node embedding methods, with promising performances on challenging tasks such as link prediction and node clustering. Graph AE, VAE and most of their extensions rely on graph convolutional networks (GCN) encoders to learn vector space representations of nodes. In this paper, we propose to replace the GCN encoder by a significantly simpler linear model w.r.t. the direct neighborhood (one-hop) adjacency matrix of the graph. For the two aforementioned tasks, we show that this approach consistently reaches competitive performances w.r.t. GCN-based models for numerous real-world graphs, including all benchmark datasets commonly used to evaluate graph AE and VAE. We question the relevance of repeatedly using these datasets to compare complex graph AE and VAE. We also emphasize the effectiveness of the proposed encoding scheme, that appears as a simpler and faster alternative to GCN encoders for many real-world applications.

使用更简单的线性模型代替多层图卷积网络编码器学习节点的向量空间表示，在节点嵌入等任务上表现良好，提出对于复杂的图自编码器和变分自编码器反复使用这些基准数据集进行比较的问题。

基于一阶线性模型的简单高效图自编码器