@article{ART003063520},
author={Jinmo Byeon and Inshil Doh and Dana Yang},
title={Enhanced ACGAN based on Progressive Step Training and Weight Transfer},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2024},
volume={29},
number={3},
pages={11-20},
doi={10.9708/jksci.2024.29.03.011}
TY - JOUR
AU - Jinmo Byeon
AU - Inshil Doh
AU - Dana Yang
TI - Enhanced ACGAN based on Progressive Step Training and Weight Transfer
JO - Journal of The Korea Society of Computer and Information
PY - 2024
VL - 29
IS - 3
PB - The Korean Society Of Computer And Information
SP - 11
EP - 20
SN - 1598-849X
AB - Among the generative models in Artificial Intelligence (AI), especially Generative Adversarial Network (GAN) has been successful in various applications such as image processing, density estimation, and style transfer. While the GAN models including Conditional GAN (CGAN), CycleGAN, BigGAN, have been extended and improved, researchers face challenges in real-world applications in specific domains such as disaster simulation, healthcare, and urban planning due to data scarcity and unstable learning causing Image distortion. This paper proposes a new progressive learning methodology called Progressive Step Training (PST) based on the Auxiliary Classifier GAN (ACGAN) that discriminates class labels, leveraging the progressive learning approach of the Progressive Growing of GAN (PGGAN). The PST model achieves 70.82% faster stabilization, 51.3% lower standard deviation, stable convergence of loss values in the later high resolution stages, and a 94.6% faster loss reduction compared to conventional methods.
KW - GAN;ACGAN;PGGAN;Step Training;Weight Adjustment;Image Generation
DO - 10.9708/jksci.2024.29.03.011
ER -
Jinmo Byeon, Inshil Doh and Dana Yang. (2024). Enhanced ACGAN based on Progressive Step Training and Weight Transfer. Journal of The Korea Society of Computer and Information, 29(3), 11-20.
Jinmo Byeon, Inshil Doh and Dana Yang. 2024, "Enhanced ACGAN based on Progressive Step Training and Weight Transfer", Journal of The Korea Society of Computer and Information, vol.29, no.3 pp.11-20. Available from: doi:10.9708/jksci.2024.29.03.011
Jinmo Byeon, Inshil Doh, Dana Yang "Enhanced ACGAN based on Progressive Step Training and Weight Transfer" Journal of The Korea Society of Computer and Information 29.3 pp.11-20 (2024) : 11.
Jinmo Byeon, Inshil Doh, Dana Yang. Enhanced ACGAN based on Progressive Step Training and Weight Transfer. 2024; 29(3), 11-20. Available from: doi:10.9708/jksci.2024.29.03.011
Jinmo Byeon, Inshil Doh and Dana Yang. "Enhanced ACGAN based on Progressive Step Training and Weight Transfer" Journal of The Korea Society of Computer and Information 29, no.3 (2024) : 11-20.doi: 10.9708/jksci.2024.29.03.011
Jinmo Byeon; Inshil Doh; Dana Yang. Enhanced ACGAN based on Progressive Step Training and Weight Transfer. Journal of The Korea Society of Computer and Information, 29(3), 11-20. doi: 10.9708/jksci.2024.29.03.011
Jinmo Byeon; Inshil Doh; Dana Yang. Enhanced ACGAN based on Progressive Step Training and Weight Transfer. Journal of The Korea Society of Computer and Information. 2024; 29(3) 11-20. doi: 10.9708/jksci.2024.29.03.011
Jinmo Byeon, Inshil Doh, Dana Yang. Enhanced ACGAN based on Progressive Step Training and Weight Transfer. 2024; 29(3), 11-20. Available from: doi:10.9708/jksci.2024.29.03.011
Jinmo Byeon, Inshil Doh and Dana Yang. "Enhanced ACGAN based on Progressive Step Training and Weight Transfer" Journal of The Korea Society of Computer and Information 29, no.3 (2024) : 11-20.doi: 10.9708/jksci.2024.29.03.011
