Generative Adversarial Networks

GANs are a class of machine learning (ML) algorithms used in unsupervised ML comprised of two deep neural networks contesting against each other (thus the word adversarial). GANs were introduced by Ian Goodfellow and other researchers at the University of Montreal, including Yoshua Bengio, in 2014.

Note

Paper on GAN by Ian Goodfellow: https://arxiv.org/abs/1406.2661.

GANs have the potential to mimic any data. This means that GANs can be trained to create similar versions of any data such as images, audio, or text. As a quick example, Christie's sold a portrait generated by GANs for $432,000, based on open source code written by Robbie Barrat of Stanford. 

A simple workflow of the GAN is shown in the following diagram:

Generative versus discriminative algorithms

To understand GANs, we should know how discriminative and generative algorithms work. Discriminative algorithms try to predict a label and classify the input data or categorize them to where the data belongs...