GAN‐based deep neural networks for graph representation learning
Abstract Graph representation learning has attracted increasing attention in a variety of applications that involve learning on non‐Euclidean data. Recently, generative adversarial networks(GAN) have been increasingly applied to the field of graph representation learning, and large progress has been...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-11-01
|
| Series: | Engineering Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/eng2.12517 |
| _version_ | 1811256396424937472 |
|---|---|
| author | Ming Zhao Yinglong Zhang |
| author_facet | Ming Zhao Yinglong Zhang |
| author_sort | Ming Zhao |
| collection | DOAJ |
| description | Abstract Graph representation learning has attracted increasing attention in a variety of applications that involve learning on non‐Euclidean data. Recently, generative adversarial networks(GAN) have been increasingly applied to the field of graph representation learning, and large progress has been made. However, most GAN‐based graph representation learning methods use adversarial learning strategies directly on the update of the vector representation instead of the deep embedding mechanism. Compared with deep models, these methods are less capable of learning nonlinear features. To address this problem and take full advantage of the essential advantages of GAN, we propose to use adversarial idea on the reconstruction mechanism of deep autoencoders. Specifically, the generator and the discriminator are the two basic components of the GAN structure. We use the deep autoencoder as the discriminator, which can capture the highly nonlinear structure of the graph. In addition, the generator another generative model is introduced into the adversarial learning system as a competitor. A series of empirical results proved the effectiveness of the new approach. |
| first_indexed | 2024-04-12T17:40:28Z |
| format | Article |
| id | doaj.art-916e5d8d1072496babbdebcee95cbe6b |
| institution | Directory Open Access Journal |
| issn | 2577-8196 |
| language | English |
| last_indexed | 2024-04-12T17:40:28Z |
| publishDate | 2022-11-01 |
| publisher | Wiley |
| record_format | Article |
| series | Engineering Reports |
| spelling | doaj.art-916e5d8d1072496babbdebcee95cbe6b2022-12-22T03:22:49ZengWileyEngineering Reports2577-81962022-11-01411n/an/a10.1002/eng2.12517GAN‐based deep neural networks for graph representation learningMing Zhao0Yinglong Zhang1School of Physics and Information Engineering Minnan Normal University Zhangzhou Fujian ChinaSchool of Physics and Information Engineering Minnan Normal University Zhangzhou Fujian ChinaAbstract Graph representation learning has attracted increasing attention in a variety of applications that involve learning on non‐Euclidean data. Recently, generative adversarial networks(GAN) have been increasingly applied to the field of graph representation learning, and large progress has been made. However, most GAN‐based graph representation learning methods use adversarial learning strategies directly on the update of the vector representation instead of the deep embedding mechanism. Compared with deep models, these methods are less capable of learning nonlinear features. To address this problem and take full advantage of the essential advantages of GAN, we propose to use adversarial idea on the reconstruction mechanism of deep autoencoders. Specifically, the generator and the discriminator are the two basic components of the GAN structure. We use the deep autoencoder as the discriminator, which can capture the highly nonlinear structure of the graph. In addition, the generator another generative model is introduced into the adversarial learning system as a competitor. A series of empirical results proved the effectiveness of the new approach.https://doi.org/10.1002/eng2.12517autoencodergenerative adversarial networkgraph representations learninggraph structure |
| spellingShingle | Ming Zhao Yinglong Zhang GAN‐based deep neural networks for graph representation learning Engineering Reports autoencoder generative adversarial network graph representations learning graph structure |
| title | GAN‐based deep neural networks for graph representation learning |
| title_full | GAN‐based deep neural networks for graph representation learning |
| title_fullStr | GAN‐based deep neural networks for graph representation learning |
| title_full_unstemmed | GAN‐based deep neural networks for graph representation learning |
| title_short | GAN‐based deep neural networks for graph representation learning |
| title_sort | gan based deep neural networks for graph representation learning |
| topic | autoencoder generative adversarial network graph representations learning graph structure |
| url | https://doi.org/10.1002/eng2.12517 |
| work_keys_str_mv | AT mingzhao ganbaseddeepneuralnetworksforgraphrepresentationlearning AT yinglongzhang ganbaseddeepneuralnetworksforgraphrepresentationlearning |