Learning Graph Topological Features via GAN
Inspired by the generation power of generative adversarial networks (GANs) in image domains, we introduce a novel hierarchical architecture for learning characteristic topological features from a single arbitrary input graph via GANs. The hierarchical architecture consisting of multiple GANs preserv...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8638941/ |
| _version_ | 1818924063207194624 |
|---|---|
| author | Weiyi Liu Pin-Yu Chen Fucai Yu Toyotaro Suzumura Guangmin Hu |
| author_facet | Weiyi Liu Pin-Yu Chen Fucai Yu Toyotaro Suzumura Guangmin Hu |
| author_sort | Weiyi Liu |
| collection | DOAJ |
| description | Inspired by the generation power of generative adversarial networks (GANs) in image domains, we introduce a novel hierarchical architecture for learning characteristic topological features from a single arbitrary input graph via GANs. The hierarchical architecture consisting of multiple GANs preserves both local and global topological features and automatically partitions the input graph into representative “stages” for feature learning. The stages facilitate reconstruction and can be used as indicators of the importance of the associated topological structures. The experiments show that our method produces subgraphs retaining a wide range of topological features, even in early reconstruction stages (unlike a single GAN, which cannot easily identify such features, let alone reconstruct the original graph). This paper is the firstline research on combining the use of GANs and graph topological analysis. |
| first_indexed | 2024-12-20T02:19:22Z |
| format | Article |
| id | doaj.art-1be3057b62fd41bfba756359ceed72c5 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T02:19:22Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-1be3057b62fd41bfba756359ceed72c52022-12-21T19:56:51ZengIEEEIEEE Access2169-35362019-01-017218342184310.1109/ACCESS.2019.28986938638941Learning Graph Topological Features via GANWeiyi Liu0https://orcid.org/0000-0003-4180-4489Pin-Yu Chen1Fucai Yu2Toyotaro Suzumura3Guangmin Hu4School of Communication and Information Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaBig Data Analytic Group, IBM Watson Research Center, Yorktown Heights, NY, USASchool of Communication and Information Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaBig Data Analytic Group, IBM Watson Research Center, Yorktown Heights, NY, USASchool of Communication and Information Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaInspired by the generation power of generative adversarial networks (GANs) in image domains, we introduce a novel hierarchical architecture for learning characteristic topological features from a single arbitrary input graph via GANs. The hierarchical architecture consisting of multiple GANs preserves both local and global topological features and automatically partitions the input graph into representative “stages” for feature learning. The stages facilitate reconstruction and can be used as indicators of the importance of the associated topological structures. The experiments show that our method produces subgraphs retaining a wide range of topological features, even in early reconstruction stages (unlike a single GAN, which cannot easily identify such features, let alone reconstruct the original graph). This paper is the firstline research on combining the use of GANs and graph topological analysis.https://ieeexplore.ieee.org/document/8638941/Generative adversarial netsgraph analysisgraph generation |
| spellingShingle | Weiyi Liu Pin-Yu Chen Fucai Yu Toyotaro Suzumura Guangmin Hu Learning Graph Topological Features via GAN IEEE Access Generative adversarial nets graph analysis graph generation |
| title | Learning Graph Topological Features via GAN |
| title_full | Learning Graph Topological Features via GAN |
| title_fullStr | Learning Graph Topological Features via GAN |
| title_full_unstemmed | Learning Graph Topological Features via GAN |
| title_short | Learning Graph Topological Features via GAN |
| title_sort | learning graph topological features via gan |
| topic | Generative adversarial nets graph analysis graph generation |
| url | https://ieeexplore.ieee.org/document/8638941/ |
| work_keys_str_mv | AT weiyiliu learninggraphtopologicalfeaturesviagan AT pinyuchen learninggraphtopologicalfeaturesviagan AT fucaiyu learninggraphtopologicalfeaturesviagan AT toyotarosuzumura learninggraphtopologicalfeaturesviagan AT guangminhu learninggraphtopologicalfeaturesviagan |