Output control of GAN model - Original GAN, derived DC-GAN, pg-GAN, etc. are "unsupervised learning" model that learns without learning labels on learning data.

Webster’s team showed that private data could be found in the output of a model; Kautz’s team showed that private data could be revealed by going in reverse, re-creating the input.

Wang used the latest GAN, developed by Nividia A.I. Labs and released by the researchers on GitHub, to train his deep learning algorithm to generate faces based on 70,000 high-resolution images.