[Updated on 2018-09-30: thanks to Yoonju, we have this post translated in Korean!] [Updated on 2019-04-18: this post is also available on arXiv.] Generative adversarial network (GAN) has shown great results in many generative tasks to replicate the real-world rich content such as images, human language, and music. It is inspired by game theory: two models, a generator and a critic, are competing with each other while making each other stronger at the same time.
From GAN to WGAN
Gradients of GAN Objectives
Read-through: Wasserstein GAN
Sensors, Free Full-Text
From GAN to WGAN
From GAN to WGAN
Comparison of the three different GAN variants: Vanilla GAN, LSGAN and
A prediction model of stock market trading actions using generative adversarial network and piecewise linear representation approaches
Calculate Value-at-Risk Using Wasserstein Generative Adversarial Networks ( WGAN-GP) for Risk Management System, by Santanu Khan