Improved Performance of GANs via Integrating Gradient Penalty with Spectral Normalization

Tan, H, Zhou, L, Wang, G and Zhang, Z 2020, Improved Performance of GANs via Integrating Gradient Penalty with Spectral Normalization, in Knowledge Science, Engineering and Management: 13th International Conference, KSEM 2020, Hangzhou, China, August 28–30, 2020, Proceedings, Part II, Springer Nature Switzerland AG, Cham, Switzerland, pp. 414-426, doi: 10.1007/978-3-030-55393-7_37.

Attached Files
Name Description MIMEType Size Downloads

Title Improved Performance of GANs via Integrating Gradient Penalty with Spectral Normalization
Author(s) Tan, H
Zhou, L
Wang, G
Zhang, ZORCID iD for Zhang, Z orcid.org/0000-0002-8721-9333
Conference name KSEM 2020
Conference location Hangzhou, China
Conference dates 2020/08/28 - 2020/08/30
Title of proceedings Knowledge Science, Engineering and Management: 13th International Conference, KSEM 2020, Hangzhou, China, August 28–30, 2020, Proceedings, Part II
Publication date 2020
Series Lecture Notes in Computer Science
Start page 414
End page 426
Total pages 13
Publisher Springer Nature Switzerland AG
Place of publication Cham, Switzerland
Keyword(s) Generative Adversarial Networks
Gradient penalty
Spectral normalization
Training stability
Networks convergence
CORE2020 B
Summary Despite the growing prominence of generative adversarial networks (GANs), improving the performance of GANs is still a challenging problem. To this end, a combination method for training GANs is proposed by coupling spectral normalization with a zero-centered gradient penalty technique (the penalty is done on the inner function of Sigmoid function of discriminator). Particularly, the proposed method not only overcomes the limitations of networks convergence and training instability but also alleviates the mode collapse behavior in GANs. Experimentally, the improved method becomes more competitive compared with some of recent methods on several datasets.
Notes Conference was held online, due to the COVID-19 pandemic.
ISBN 9783030553920
ISSN 0302-9743
1611-3349
Language eng
DOI 10.1007/978-3-030-55393-7_37
Indigenous content off
HERDC Research category E1 Full written paper - refereed
Copyright notice ©2020, Springer Nature Switzerland AG
Persistent URL http://hdl.handle.net/10536/DRO/DU:30141950

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 0 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 34 Abstract Views, 4 File Downloads  -  Detailed Statistics
Created: Wed, 16 Sep 2020, 09:42:06 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.