Poster

Tailoring Self-Attention for Graph via Rooted Subtrees

Siyuan Huang · Yunchong Song · Jiayue Zhou · Zhouhan Lin

Great Hall & Hall B1+B2 (level 1) #818
[ ] [ Project Page ]
Thu 14 Dec 3 p.m. PST — 5 p.m. PST

Abstract:

Attention mechanisms have made significant strides in graph learning, yet they still exhibit notable limitations: local attention faces challenges in capturing long-range information due to the inherent problems of the message-passing scheme, while global attention cannot reflect the hierarchical neighborhood structure and fails to capture fine-grained local information. In this paper, we propose a novel multi-hop graph attention mechanism, named Subtree Attention (STA), to address the aforementioned issues. STA seamlessly bridges the fully-attentional structure and the rooted subtree, with theoretical proof that STA approximates the global attention under extreme settings. By allowing direct computation of attention weights among multi-hop neighbors, STA mitigates the inherent problems in existing graph attention mechanisms. Further we devise an efficient form for STA by employing kernelized softmax, which yields a linear time complexity. Our resulting GNN architecture, the STAGNN, presents a simple yet performant STA-based graph neural network leveraging a hop-aware attention strategy. Comprehensive evaluations on ten node classification datasets demonstrate that STA-based models outperform existing graph transformers and mainstream GNNs. The codeis available at https://github.com/LUMIA-Group/SubTree-Attention.

Chat is not available.