HAT-GAE: Self-Supervised Graph Auto-encoders with Hierarchical Adaptive Masking and Trainable Corruption

TL;DR

This paper introduces HAT-GAE, a novel self-supervised graph auto-encoder with hierarchical adaptive masking and trainable corruption, improving graph representation learning by mimicking human learning and enhancing robustness.

Contribution

The paper proposes a new auto-encoder architecture for graphs that uses hierarchical adaptive masking and trainable corruption to boost learning and robustness.

Findings

Outperforms state-of-the-art models on ten benchmark datasets.

Demonstrates improved robustness and learning efficiency.

Validates effectiveness of hierarchical adaptive masking.

Abstract

Self-supervised auto-encoders have emerged as a successful framework for representation learning in computer vision and natural language processing in recent years, However, their application to graph data has been met with limited performance due to the non-Euclidean and complex structure of graphs in comparison to images or text, as well as the limitations of conventional auto-encoder architectures. In this paper, we investigate factors impacting the performance of auto-encoders on graph data and propose a novel auto-encoder model for graph representation learning. Our model incorporates a hierarchical adaptive masking mechanism to incrementally increase the difficulty of training in order to mimic the process of human cognitive learning, and a trainable corruption scheme to enhance the robustness of learned representations. Through extensive experimentation on ten benchmark datasets, we demonstrate the superiority of our proposed method over state-of-the-art graph representation learning models.