Adversarial net and its variants

Changan Du; Yukun Huang; Huajie Zeng

doi:10.1117/12.2639238

17 May 2022 Adversarial net and its variants

Changan Du, Yukun Huang, Huajie Zeng

Proceedings Volume 12259, 2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2022); 122595H (2022) https://doi.org/10.1117/12.2639238
Event: 2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing, 2022, Kunming, China

Abstract

With the rapid development of computer science, generative adversarial networks (GAN) have become increasingly important in machine learning. GAN is a machine learning model consisting of a generative network and discriminative network and it is a two-player game proposed by Goodfellow. GAN plays a significant role in real-life application, especially in the image to image translation, image synthesis, data augmentation, and image editing. However, GANs have several issues that have not been well addressed: non-convergence problem, collapse problem, and so on. This article provides a comprehensive overview and analysis of GAN. Firstly, the theory, architecture, and training procedure of the original GAN are introduced in detail. Moreover, we present several variants of GAN, including DCGAN, BEGAN, LSGAN, CIAGAN, and WGAN. We discuss the motivation and networks structures of these variants and list and summarize their advantages and difference. Finally, we provide readers with several evaluation metrics for GAN and point out future open research challenges.

Citation Download Citation

Changan Du, Yukun Huang, and Huajie Zeng "Adversarial net and its variants", Proc. SPIE 12259, 2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2022), 122595H (17 May 2022); https://doi.org/10.1117/12.2639238

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Gallium nitride

Network architectures

Machine learning

Statistical modeling

Data modeling

Error analysis

Image quality

Show All Keywords

Keywords/Phrases

Search In:

Publication Years