BriefGPT.xyz
Feb, 2019
提高生成对抗网络的泛化能力和稳定性
Improving Generalization and Stability of Generative Adversarial Networks
HTML
PDF
Hoang Thanh-Tung, Truyen Tran, Svetha Venkatesh
TL;DR
本文通过分析实际情景下GAN的泛化能力，证明了原始GAN的损失函数训练得到的鉴别器的泛化能力较差，并提出了一种零中心梯度惩罚策略以改善鉴别器的泛化能力，并保证GAN的收敛和泛化。通过在合成和大规模数据集上的实验，验证了理论分析的正确性。
Abstract
generative adversarial networks
(GANs) are one of the most popular tools for learning complex high dimensional distributions. However,
generalization
properties of GANs have not been well understood. In this pape
→