BriefGPT.xyz
Jun, 2018
生成对抗网络（GANs）的收敛问题
Convergence Problems with Generative Adversarial Networks (GANs)
HTML
PDF
Samuel A. Barnett
TL;DR
本文旨在为数学家提供适用的GANs理论解释，概述GANs的训练问题和拓扑学和博弈论视角如何贡献于我们理解和实践GANs的技术的正面和反面结果。
Abstract
generative adversarial networks
(
gans
) are a novel approach to generative modelling, a task whose goal it is to learn a distribution of real data points. They have often proved difficult to train:
→