.png){kind=icon}
.png){kind=icon}
Return
Architecture Knowledge Distillation for Evolutionary Generative Adversarial Network
.png){kind=icon}
PRE
AI
DOI：10.1142/S0129065725500133.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Generative Adversarial Networks (GANs) are effective for image generation, but their unstable training limits broader applications. Additionally, neural architecture search (NAS) for GANs with one-shot models often leads to insufficient subnet training, where subnets inherit weights from a supernet without proper optimization, further degrading performance. To address both issues, we propose Architecture Knowledge Distillation for Evolutionary GAN (AKD-EGAN). AKD-EGAN operates in two stages. First, architecture knowledge distillation (AKD) is used during supernet training to efficiently optimize subnetworks and accelerate learning. Second, a multi-objective evolutionary algorithm (MOEA) searches for optimal subnet architectures, ensuring efficiency by considering multiple performance metrics. This approach, combined with a strategy for architecture inheritance, enhances GAN stability and image quality. Experiments show that AKD-EGAN surpasses state-of-the-art methods, achieving a Fr & eacute;chet Inception Distance (FID) of 7.91 and an Inception Score (IS) of 8.97 on CIFAR-10, along with competitive results on STL-10 (FID: 20.32, IS: 10.06). Code and models will be available at https://github.com/njit-ly/AKD-EGAN.
Keywords:
Neural architecture search
generative adversarial network
evolutionary computation
architecture knowledge distillation
generative model
Journal
IF:
6.4
Papers:
1.2K
Citations:
3.3K