Global Attention Improves Graph Networks Generalization

Omri Puny, Heli Ben-Hamu, Yaron Lipman

Keywords: Graph Neural Network, Self-Attention, Generalization of GNNs, Weisfeiler-Lehman

Abstract:

This paper advocates incorporating a Low-Rank Global Attention (LRGA) module, a Computation and memory efficient variant of the dot-product attention (Vaswani et al., 2017), to Graph Neural Networks (GNNs) for improving their generalization power.
To theoretically quantify the generalization properties granted by adding the LRGA module to GNNs, we focus on a specific family of expressive GNNs and show that augmenting it with LRGA provides algorithmic alignment to a powerful graph isomorphism test, namely the 2-Folklore Weisfeiler-Lehman (2-FWL) algorithm. In more detail we: (i) consider the recent Random Graph Neural Network (RGNN) (Sato et al., 2020) framework and prove that it is universal in probability; (ii) show that RGNN augmented with LRGA aligns with 2-FWL update step via polynomial kernels; and (iii) bound the sample complexity of the kernel's feature map when learned with a randomly initialized two-layer MLP.
From a practical point of view, augmenting existing GNN layers with LRGA produces state of the art results in current GNN benchmarks. Lastly, we observe that augmenting various GNN architectures with LRGA often closes the performance gap across different models.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

One-sentence Summary: Low rank global self-attention for improving generalization of graph neural networks

Reviewed Version (pdf): https://openreview.net/references/pdf?id=SalR_R6uny