International Transactions on Data Science, Engineering and Technology

Manuscript In Press (Unedited Version)

Research progress of generative adversarial networks GAN | Review Article

Yongqiang Zhang a,, Xiaohan Sun b, Wanzhen Zhou c, Menghua Man d
a College of Information Science and Engineering，Hebei University of Science and Technology, ShiJiazhuang, China
b College of Information Science and Engineering，Hebei University of Science and Technology, ShiJiazhuang, China
c College of Information Science and Engineering，Hebei University of Science and Technology, ShiJiazhuang, China
d Shijiazhuang Campus of Army Engineering University, ShiJiazhuang, China

Highlights and Novelties


This article first introduces the basic theory, loss function and network model of GAN, starting from the traditional GAN algorithm, summarizes the more prominent aspects of GAN research in recent years

Then it analyzes and summarizes the derivative model of GAN in detail, and introduces several common evaluation indexes of GAN

Summarized the practical application of GAN in the field of vision, natural language processing, and various other fields.


Manuscript Abstract
Machine learning is an important field in artificial intelligence research. It is divided into two types: supervised learning and unsupervised learning. As an unsupervised learning method, generative adversarial network has been proposed in 2014, and it has been studied by many scholars. It trains the neural network through mutual adversarial learning, which makes the training and generation results of the model more convergent. This article first introduces the basic theory, loss function and network model of GAN, starting from the traditional GAN algorithm, summarizes the more prominent aspects of GAN research in recent years. Then it analyzes and summarizes the derivative model of GAN in detail, and introduces several common evaluation indexes of GAN. Summarized the practical application of GAN in the field of vision, natural language processing, and various other fields. Finally, it summarizes the current problems of GAN, and looks forward to the future development trend of GAN.

Keywords
 Generative Adversarial Network   Adversarial Learning   Unsupervised Learning   Neural Network 

Copyright
© Copyright was transferred to International Computer Science and Engineering Society (ICSES) by all the Authors.

Cite this manuscript as
Yongqiang Zhang, Xiaohan Sun, Wanzhen Zhou, Menghua Man, "Research progress of generative adversarial networks GAN," International Transactions on Data Science, Engineering and Technology (ITDSET), In Press, pp. 1-7, Sep. 2021.

For External Scientific Databeses
--BibTex-- --EndNote-- --Dublin--
Written by: Admin | Link ... |