@article{ART003148627},
author={Seung-Yeon Lee and Heo seok yeol and Lee wan jik},
title={Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning},
journal={Journal of Software Assessment and Valuation},
issn={2092-8114},
year={2024},
volume={20},
number={4},
pages={145-154}
TY - JOUR
AU - Seung-Yeon Lee
AU - Heo seok yeol
AU - Lee wan jik
TI - Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning
JO - Journal of Software Assessment and Valuation
PY - 2024
VL - 20
IS - 4
PB - Korea Software Assessment and Valuation Society
SP - 145
EP - 154
SN - 2092-8114
AB - Object recognition technology is widely used as a basic technology for process automation, autonomous driving, and smart system construction in the field of artificial intelligence. In such object recognition technology, recursive learning techniques are widely used to improve the accuracy of recognition, in which images generated by artificial intelligence are reused as learning data. However, this recursive learning can cause the problem of model collapse, which reduces the performance of object recognition. To solve this problem, in this paper, we mathematically analyzes the image quality degradation process that causes model collapse, and propose a generated image evaluation technique using singularity analysis and natural logarithm function based on this. In addition, the image evaluation technique in the paper was experimentally implemented, and analysis was performed on images with different damage ratios from the original image. Through these experiments and experimental results, it was shown that the image evaluation technique proposed in the paper can be sufficiently utilized to improve the model collapse phenomenon by early detecting image damage that can occur in model collapse of recursive learning.
KW - Artificial Intelligence;Recursive Learning;Model Collapse Improvement;Singularity Analysis;Image Evaluation Techniques
DO -
UR -
ER -
Seung-Yeon Lee, Heo seok yeol and Lee wan jik. (2024). Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning. Journal of Software Assessment and Valuation, 20(4), 145-154.
Seung-Yeon Lee, Heo seok yeol and Lee wan jik. 2024, "Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning", Journal of Software Assessment and Valuation, vol.20, no.4 pp.145-154.
Seung-Yeon Lee, Heo seok yeol, Lee wan jik "Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning" Journal of Software Assessment and Valuation 20.4 pp.145-154 (2024) : 145.
Seung-Yeon Lee, Heo seok yeol, Lee wan jik. Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning. 2024; 20(4), 145-154.
Seung-Yeon Lee, Heo seok yeol and Lee wan jik. "Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning" Journal of Software Assessment and Valuation 20, no.4 (2024) : 145-154.
Seung-Yeon Lee; Heo seok yeol; Lee wan jik. Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning. Journal of Software Assessment and Valuation, 20(4), 145-154.
Seung-Yeon Lee; Heo seok yeol; Lee wan jik. Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning. Journal of Software Assessment and Valuation. 2024; 20(4) 145-154.
Seung-Yeon Lee, Heo seok yeol, Lee wan jik. Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning. 2024; 20(4), 145-154.
Seung-Yeon Lee, Heo seok yeol and Lee wan jik. "Improvement of Model Collapse Phenomenon due to Artificial Intelligence Recursive Learning" Journal of Software Assessment and Valuation 20, no.4 (2024) : 145-154.
