Focus Where It Matters: Graph Selective State Focused Attention Networks

TL;DR

This paper introduces GSAN, a novel graph neural network architecture that uses multi-head masked self-attention and selective state space modeling to improve scalability, adaptability, and performance on dynamic graph tasks.

Contribution

The paper proposes GSAN, combining MHMSA and S3M layers, to address over-smoothing and scalability issues in deep GNNs, enhancing dynamic graph learning and interpretability.

Findings

GSAN outperforms traditional GNNs on benchmark datasets.

GSAN improves classification accuracy by up to 8.94%.

The model effectively handles evolving graph environments.

Abstract

Traditional graph neural networks (GNNs) lack scalability and lose individual node characteristics due to over-smoothing, especially in the case of deeper networks. This results in sub-optimal feature representation, affecting the model's performance on tasks involving dynamically changing graphs. To address this issue, we present Graph Selective States Focused Attention Networks (GSANs) based neural network architecture for graph-structured data. The GSAN is enabled by multi-head masked self-attention (MHMSA) and selective state space modeling (S3M) layers to overcome the limitations of GNNs. In GSAN, the MHMSA allows GSAN to dynamically emphasize crucial node connections, particularly in evolving graph environments. The S3M layer enables the network to adjust dynamically in changing node states and improving predictions of node behavior in varying contexts without needing primary knowledge of the graph structure. Furthermore, the S3M layer enhances the generalization of unseen structures and interprets how node states influence link importance. With this, GSAN effectively outperforms inductive and transductive tasks and overcomes the issues that traditional GNNs experience. To analyze the performance behavior of GSAN, a set of state-of-the-art comparative experiments are conducted on graphs benchmark datasets, including $Cora$, $Citeseer$, $Pubmed$ network citation, and $protein-protein-interaction$ datasets, as an outcome, GSAN improved the classification accuracy by $1.56\%$, $8.94\%$, $0.37\%$, and $1.54\%$ on $F1-score$ respectively.