BriefGPT.xyz
Sep, 2020
TinyGAN:基于BigGAN的有条件图像生成蒸馏
TinyGAN: Distilling BigGAN for Conditional Image Generation
HTML
PDF
Ting-Yun Chang, Chi-Jen Lu
TL;DR
提出了一种基于黑盒知识蒸馏的GAN模型压缩方法,将BigGAN作为教师网络,用较少的参数训练学生网络以模拟其功能,有效地缩小模型规模并在图像生成方面取得了有竞争力的表现。
Abstract
generative adversarial networks
(GANs) have become a powerful approach for generative image modeling. However, GANs are notorious for their training instability, especially on large-scale, complex datasets. While the recent work of
→