Neural Network Programming with TensorFlow

Vanilla GAN

Vanilla GANs has two networks called generator network and a discriminator network. Both the networks are trained at the same time and compete or battle against each other in a minimax play. Generator network is trained or prepared such that it can fool the discriminator network by creating the real images as per the input, and the discriminator is trained not to be fooled by the generator network.

For further reading on Vanilla GAN refer to https://arxiv.org/abs/1406.2661.