Quantum-based subgraph convolutional neural networks

TY - JOUR

T1 - Quantum-based subgraph convolutional neural networks

AU - Zhang, Zhihong

AU - Chen, Dongdong

AU - Wang, Jianjia

AU - Bai, Lu

AU - Hancock, Edwin R.

N1 - © 2018 Elsevier Ltd. All rights reserved.This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.

PY - 2018/11/6

Y1 - 2018/11/6

N2 - This paper proposes a new graph convolutional neural network architecture based on a depth-based representation of graph structure deriving from quantum walks, which we refer to as the quantum-based subgraph convolutional neural network (QS-CNNs). This new architecture captures both the global topological structure and the local connectivity structure within a graph. Specifically, we commence by establishing a family of K-layer expansion subgraphs for each vertex of a graph by quantum walks, which captures the global topological arrangement information for substructures contained within a graph. We then design a set of fixed-size convolution filters over the subgraphs, which helps to characterise multi-scale patterns residing in the data. The idea is to apply convolution filters sliding over the entire set of subgraphs rooted at a vertex to extract the local features analogous to the standard convolution operation on grid data. Experiments on eight graph-structured datasets demonstrate that QS-CNNs architecture is capable of outperforming fourteen state-of-the-art methods for the tasks of node classification and graph classification.

AB - This paper proposes a new graph convolutional neural network architecture based on a depth-based representation of graph structure deriving from quantum walks, which we refer to as the quantum-based subgraph convolutional neural network (QS-CNNs). This new architecture captures both the global topological structure and the local connectivity structure within a graph. Specifically, we commence by establishing a family of K-layer expansion subgraphs for each vertex of a graph by quantum walks, which captures the global topological arrangement information for substructures contained within a graph. We then design a set of fixed-size convolution filters over the subgraphs, which helps to characterise multi-scale patterns residing in the data. The idea is to apply convolution filters sliding over the entire set of subgraphs rooted at a vertex to extract the local features analogous to the standard convolution operation on grid data. Experiments on eight graph-structured datasets demonstrate that QS-CNNs architecture is capable of outperforming fourteen state-of-the-art methods for the tasks of node classification and graph classification.

KW - Graph convolutional neural networks, Spatial construction, Quantum walks, Subgraph

U2 - 10.1016/j.patcog.2018.11.002

DO - 10.1016/j.patcog.2018.11.002

M3 - Article

VL - 88

SP - 38

EP - 49

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

ER -