Research output: Working paper › Preprint

**Graph Convolutional Neural Networks based on Quantum Vertex Saliency.** / Bai, Lu; Jiao, Yuhang; Rossi, Luca; Cui, Lixin; Cheng, Jian; Hancock, Edwin R.

Research output: Working paper › Preprint

Bai, L, Jiao, Y, Rossi, L, Cui, L, Cheng, J & Hancock, ER 2018 'Graph Convolutional Neural Networks based on Quantum Vertex Saliency' arXiv. <https://arxiv.org/abs/1809.01090>

Bai, L., Jiao, Y., Rossi, L., Cui, L., Cheng, J., & Hancock, E. R. (2018). *Graph Convolutional Neural Networks based on Quantum Vertex Saliency*. (arXiv). https://arxiv.org/abs/1809.01090

Bai L, Jiao Y, Rossi L, Cui L, Cheng J, Hancock ER. Graph Convolutional Neural Networks based on Quantum Vertex Saliency. 2018 Sep 4. (arXiv).

@techreport{a3dc89ecdd2d4579a55dd7df94487708,

title = "Graph Convolutional Neural Networks based on Quantum Vertex Saliency",

abstract = " This paper proposes a new Quantum Spatial Graph Convolutional Neural Network (QSGCNN) model that can directly learn a classification function for graphs of arbitrary sizes. Unlike state-of-the-art Graph Convolutional Neural Network (GCNN) models, the proposed QSGCNN model incorporates the process of identifying transitive aligned vertices between graphs, and transforms arbitrary sized graphs into fixed-sized aligned vertex grid structures. In order to learn representative graph characteristics, a new quantum spatial graph convolution is proposed and employed to extract multi-scale vertex features, in terms of quantum information propagation between grid vertices of each graph. Since the quantum spatial convolution preserves the grid structures of the input vertices (i.e., the convolution layer does not change the original spatial sequence of vertices), the proposed QSGCNN model allows to directly employ the traditional convolutional neural network architecture to further learn from the global graph topology, providing an end-to-end deep learning architecture that integrates the graph representation and learning in the quantum spatial graph convolution layer and the traditional convolutional layer for graph classifications. We demonstrate the effectiveness of the proposed QSGCNN model in relation to existing state-of-the-art methods. The proposed QSGCNN model addresses the shortcomings of information loss and imprecise information representation arising in existing GCN models associated with the use of SortPooling or SumPooling layers. Experiments on benchmark graph classification datasets demonstrate the effectiveness of the proposed QSGCNN model. ",

keywords = "cs.LG, stat.ML",

author = "Lu Bai and Yuhang Jiao and Luca Rossi and Lixin Cui and Jian Cheng and Hancock, {Edwin R.}",

year = "2018",

month = sep,

day = "4",

language = "English",

series = "arXiv",

type = "WorkingPaper",

}

TY - UNPB

T1 - Graph Convolutional Neural Networks based on Quantum Vertex Saliency

AU - Bai, Lu

AU - Jiao, Yuhang

AU - Rossi, Luca

AU - Cui, Lixin

AU - Cheng, Jian

AU - Hancock, Edwin R.

PY - 2018/9/4

Y1 - 2018/9/4

N2 - This paper proposes a new Quantum Spatial Graph Convolutional Neural Network (QSGCNN) model that can directly learn a classification function for graphs of arbitrary sizes. Unlike state-of-the-art Graph Convolutional Neural Network (GCNN) models, the proposed QSGCNN model incorporates the process of identifying transitive aligned vertices between graphs, and transforms arbitrary sized graphs into fixed-sized aligned vertex grid structures. In order to learn representative graph characteristics, a new quantum spatial graph convolution is proposed and employed to extract multi-scale vertex features, in terms of quantum information propagation between grid vertices of each graph. Since the quantum spatial convolution preserves the grid structures of the input vertices (i.e., the convolution layer does not change the original spatial sequence of vertices), the proposed QSGCNN model allows to directly employ the traditional convolutional neural network architecture to further learn from the global graph topology, providing an end-to-end deep learning architecture that integrates the graph representation and learning in the quantum spatial graph convolution layer and the traditional convolutional layer for graph classifications. We demonstrate the effectiveness of the proposed QSGCNN model in relation to existing state-of-the-art methods. The proposed QSGCNN model addresses the shortcomings of information loss and imprecise information representation arising in existing GCN models associated with the use of SortPooling or SumPooling layers. Experiments on benchmark graph classification datasets demonstrate the effectiveness of the proposed QSGCNN model.

AB - This paper proposes a new Quantum Spatial Graph Convolutional Neural Network (QSGCNN) model that can directly learn a classification function for graphs of arbitrary sizes. Unlike state-of-the-art Graph Convolutional Neural Network (GCNN) models, the proposed QSGCNN model incorporates the process of identifying transitive aligned vertices between graphs, and transforms arbitrary sized graphs into fixed-sized aligned vertex grid structures. In order to learn representative graph characteristics, a new quantum spatial graph convolution is proposed and employed to extract multi-scale vertex features, in terms of quantum information propagation between grid vertices of each graph. Since the quantum spatial convolution preserves the grid structures of the input vertices (i.e., the convolution layer does not change the original spatial sequence of vertices), the proposed QSGCNN model allows to directly employ the traditional convolutional neural network architecture to further learn from the global graph topology, providing an end-to-end deep learning architecture that integrates the graph representation and learning in the quantum spatial graph convolution layer and the traditional convolutional layer for graph classifications. We demonstrate the effectiveness of the proposed QSGCNN model in relation to existing state-of-the-art methods. The proposed QSGCNN model addresses the shortcomings of information loss and imprecise information representation arising in existing GCN models associated with the use of SortPooling or SumPooling layers. Experiments on benchmark graph classification datasets demonstrate the effectiveness of the proposed QSGCNN model.

KW - cs.LG

KW - stat.ML

M3 - Preprint

T3 - arXiv

BT - Graph Convolutional Neural Networks based on Quantum Vertex Saliency

ER -