Generative Adversarial Network (GAN) in Deep Learning

In this video, we will learn about GAN which is also known as Generative Adversarial Neural Networks and an unsupervised machine learning algorithm.

A GAN model has two major parts:
Generator - Creates a fake image from noise
Discriminator - Discriminate between the fake and the real image

Our objective is to improve the generator upto a such a level so, that the discriminator part of the model becomes unable to classify between the real and the fake images. 

Nowadays GAN’s have many application like motionizing a static image. But the main purpose for which GAN’s were build initially was to create fake data to solve the problem of scarcity of data for some cases.

Generative adversarial network:
https://www.geeksforgeeks.org/generative-adversarial-networks-gans-an-introduction/
https://www.geeksforgeeks.org/generative-adversarial-network-gan/