We present masked graph autoencoder (MaskGAE), a self-supervised learning
framework for graph-structured data. Different from previous graph autoencoders
(GAEs), MaskGAE adopts masked graph modeling (MGM) as a principled pretext
task: masking a portion of edges and attempting to reconstruct the missing part
with partially visible, unmasked graph structure. To understand whether MGM can
help GAEs learn better representations, we provide both theoretical and
empirical evidence to justify the benefits of this pretext task. Theoretically,
we establish the connections between GAEs and contrastive learning, showing
that MGM significantly improves the self-supervised learning scheme of GAEs.
Empirically, we conduct extensive experiments on a number of benchmark
datasets, demonstrating the superiority of MaskGAE over several
state-of-the-arts on both link prediction and node classification tasks. Our
code is publicly available at https://github.com/EdisonLeeeee/MaskGAE.

本文提出了自监督学习框架 MaskGAE，以 Masked Graph Modeling 为前置任务，证明其可以大大提高自监督学习方案。经过广泛的基准测试，证明了 MaskGAE 在连接预测和节点分类任务上的优越性。