The Community for Technology Leaders
2018 24th International Conference on Pattern Recognition (ICPR) (2018)
Beijing, China
Aug. 20, 2018 to Aug. 24, 2018
ISSN: 1051-4651
ISBN: 978-1-5386-3789-0
pp: 1024-1029
Chuanyu Xu , Xiamen University, Xiamen, China
Dong Wang , Xiamen University, Xiamen, China
Zhihong Zhang , Xiamen University, Xiamen, China
Beizhan Wang , Xiamen University, Xiamen, China
Da Zhou , Xiamen University, Xiamen, China
Guijun Ren , Opera Solutions, LLC, Capital Markets Analytics, Jersey City, NJ, 07302, USA
Lu Bai , Department of Computer Science, University of York, York, UK
Lixin Cui , Central University of Finance and Economics, Beijing, China
Edwin R. Hancock , Department of Computer Science, University of York, York, UK
ABSTRACT
This paper proposes a new graph convolutional neural architecture based on a depth-based representation of graph structure, called the depth-based subgraph convolutional neural networks (DS-CNNs), which integrates both the global topological and local connectivity structures within a graph. Our idea is to decompose a graph into a family of $K$-layer expansion subgraphs rooted at each vertex, and then a set of convolution filters are designed over these subgraphs to capture local connectivity structural information. Specifically, we commence by establishing a family of $K$-layer expansion subgraphs for each vertex of graph by mapping graph to tree procedures, which can provide global topological arrangement information contained within a graph. We then design a set of fixed-size convolution filters and integrate them with these subgraphs (depicted in Figure 1). The idea is to apply convolution filters sliding over the entire subgraphs of a vertex to extract the local features analogous to the standard convolution operation on grid data. In particular, the convolution operation captures the local structural information within the graph, and has the weight sharing property among different positions of subgraph; the pooling operation acts directly on the output of the preceding layer without any preprocessing scheme (e.g., clustering or other techniques). Experiments on three graph-structured datasets demonstrate that our model DS-CNNs are able to outperform six state-of-the-art methods at the task of node classification.
INDEX TERMS
Convolution, Feature extraction, Convolutional neural networks, Standards, Task analysis, Data mining, Laplace equations
CITATION

C. Xu et al., "Depth-based Subgraph Convolutional Neural Networks," 2018 24th International Conference on Pattern Recognition (ICPR), Beijing, China, 2018, pp. 1024-1029.
doi:10.1109/ICPR.2018.8545090
915 ms
(Ver 3.3 (11022016))