### Abstract

Graph data widely exist in many high-impact applications. Inspired by the success of deep learning in grid-structured data, graph neural network models have been proposed to learn powerful node-level or graph-level representation. However, most of the existing graph neural networks suffer from the following limitations: (1) there is limited analysis regarding the graph convolution properties, such as seed-oriented, degree-aware and order-free; (2) the node's degree-specific graph structure is not explicitly expressed in graph convolution for distinguishing structure-aware node neighborhoods; (3) the theoretical explanation regarding the graph-level pooling schemes is unclear. To address these problems, we propose a generic degree-specific graph neural network named DEMO-Net motivated by Weisfeiler-Lehman graph isomorphism test that recursively identifies 1-hop neighborhood structures. In order to explicitly capture the graph topology integrated with node attributes, we argue that graph convolution should have three properties: seed-oriented, degree-aware, order-free. To this end, we propose multi-task graph convolution where each task represents node representation learning for nodes with a specific degree value, thus leading to preserving the degree-specific graph structure. In particular, we design two multi-task learning methods: degree-specific weight and hashing functions for graph convolution. In addition, we propose a novel graph-level pooling/readout scheme for learning graph representation provably lying in a degree-specific Hilbert kernel space. The experimental results on several node and graph classification benchmark data sets demonstrate the effectiveness and efficiency of our proposed DEMO-Net over state-of-the-art graph neural network models.

Original language | English (US) |
---|---|

Title of host publication | KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining |

Publisher | Association for Computing Machinery |

Pages | 406-415 |

Number of pages | 10 |

ISBN (Electronic) | 9781450362016 |

DOIs | |

State | Published - Jul 25 2019 |

Event | 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019 - Anchorage, United States Duration: Aug 4 2019 → Aug 8 2019 |

### Publication series

Name | Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining |
---|

### Conference

Conference | 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019 |
---|---|

Country | United States |

City | Anchorage |

Period | 8/4/19 → 8/8/19 |

### Fingerprint

### Keywords

- Degree-specific Convolution
- Graph Isomorphism Test
- Graph Neural Network
- Multi-task Learning

### ASJC Scopus subject areas

- Software
- Information Systems

### Cite this

*KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining*(pp. 406-415). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3292500.3330950

**Demo-Net : Degree-specific graph neural networks for node and graph classification.** / Wu, Jun; He, Jingrui; Xu, Jiejun.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.*Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 406-415, 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2019, Anchorage, United States, 8/4/19. https://doi.org/10.1145/3292500.3330950

}

TY - GEN

T1 - Demo-Net

T2 - Degree-specific graph neural networks for node and graph classification

AU - Wu, Jun

AU - He, Jingrui

AU - Xu, Jiejun

PY - 2019/7/25

Y1 - 2019/7/25

N2 - Graph data widely exist in many high-impact applications. Inspired by the success of deep learning in grid-structured data, graph neural network models have been proposed to learn powerful node-level or graph-level representation. However, most of the existing graph neural networks suffer from the following limitations: (1) there is limited analysis regarding the graph convolution properties, such as seed-oriented, degree-aware and order-free; (2) the node's degree-specific graph structure is not explicitly expressed in graph convolution for distinguishing structure-aware node neighborhoods; (3) the theoretical explanation regarding the graph-level pooling schemes is unclear. To address these problems, we propose a generic degree-specific graph neural network named DEMO-Net motivated by Weisfeiler-Lehman graph isomorphism test that recursively identifies 1-hop neighborhood structures. In order to explicitly capture the graph topology integrated with node attributes, we argue that graph convolution should have three properties: seed-oriented, degree-aware, order-free. To this end, we propose multi-task graph convolution where each task represents node representation learning for nodes with a specific degree value, thus leading to preserving the degree-specific graph structure. In particular, we design two multi-task learning methods: degree-specific weight and hashing functions for graph convolution. In addition, we propose a novel graph-level pooling/readout scheme for learning graph representation provably lying in a degree-specific Hilbert kernel space. The experimental results on several node and graph classification benchmark data sets demonstrate the effectiveness and efficiency of our proposed DEMO-Net over state-of-the-art graph neural network models.

AB - Graph data widely exist in many high-impact applications. Inspired by the success of deep learning in grid-structured data, graph neural network models have been proposed to learn powerful node-level or graph-level representation. However, most of the existing graph neural networks suffer from the following limitations: (1) there is limited analysis regarding the graph convolution properties, such as seed-oriented, degree-aware and order-free; (2) the node's degree-specific graph structure is not explicitly expressed in graph convolution for distinguishing structure-aware node neighborhoods; (3) the theoretical explanation regarding the graph-level pooling schemes is unclear. To address these problems, we propose a generic degree-specific graph neural network named DEMO-Net motivated by Weisfeiler-Lehman graph isomorphism test that recursively identifies 1-hop neighborhood structures. In order to explicitly capture the graph topology integrated with node attributes, we argue that graph convolution should have three properties: seed-oriented, degree-aware, order-free. To this end, we propose multi-task graph convolution where each task represents node representation learning for nodes with a specific degree value, thus leading to preserving the degree-specific graph structure. In particular, we design two multi-task learning methods: degree-specific weight and hashing functions for graph convolution. In addition, we propose a novel graph-level pooling/readout scheme for learning graph representation provably lying in a degree-specific Hilbert kernel space. The experimental results on several node and graph classification benchmark data sets demonstrate the effectiveness and efficiency of our proposed DEMO-Net over state-of-the-art graph neural network models.

KW - Degree-specific Convolution

KW - Graph Isomorphism Test

KW - Graph Neural Network

KW - Multi-task Learning

UR - http://www.scopus.com/inward/record.url?scp=85071188809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071188809&partnerID=8YFLogxK

U2 - 10.1145/3292500.3330950

DO - 10.1145/3292500.3330950

M3 - Conference contribution

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 406

EP - 415

BT - KDD 2019 - Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -