.png){kind=icon}
.png){kind=icon}
Return
GEGAN: gradient-guided evolutionary framework for GAN optimization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2026.131257.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes a gradient-guided mutation operator to enhance GAN output quality and diversity. • Utilizes gradient information to accelerate generator convergence and improve training stability. • Develops GEGAN with proven convergence under general conditions and an acceptance-based strategy. • Empirically validates GEGAN’s superiority in data-scarce, multi-class, and noisy environments.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W