Cluster-wise Graph Transformer with Dual-granularity Kernelized Attention

TL;DR

This paper introduces a novel graph transformer architecture that captures dual-granularity information through a cluster-wise attention mechanism, improving performance on graph-level tasks without losing node-level details.

Contribution

It proposes the Node-to-Cluster Attention (N2C-Attn) mechanism with kernelized attention, enabling efficient dual-granularity information transfer in graph transformers.

Findings

Achieves linear time complexity with the cluster-wise message-passing framework.

Demonstrates superior performance on various graph-level tasks.

Effectively merges node and cluster-level features using dual-granularity attention.

Abstract

In the realm of graph learning, there is a category of methods that conceptualize graphs as hierarchical structures, utilizing node clustering to capture broader structural information. While generally effective, these methods often rely on a fixed graph coarsening routine, leading to overly homogeneous cluster representations and loss of node-level information. In this paper, we envision the graph as a network of interconnected node sets without compressing each cluster into a single embedding. To enable effective information transfer among these node sets, we propose the Node-to-Cluster Attention (N2C-Attn) mechanism. N2C-Attn incorporates techniques from Multiple Kernel Learning into the kernelized attention framework, effectively capturing information at both node and cluster levels. We then devise an efficient form for N2C-Attn using the cluster-wise message-passing framework, achieving linear time complexity. We further analyze how N2C-Attn combines bi-level feature maps of queries and keys, demonstrating its capability to merge dual-granularity information. The resulting architecture, Cluster-wise Graph Transformer (Cluster-GT), which uses node clusters as tokens and employs our proposed N2C-Attn module, shows superior performance on various graph-level tasks. Code is available at https://github.com/LUMIA-Group/Cluster-wise-Graph-Transformer.