# Difference between revisions of "stat946w18/IMPROVING GANS USING OPTIMAL TRANSPORT"

(→GANs AND OPTIMAL TRANSPORT) |
(→GANs AND OPTIMAL TRANSPORT) |
||

Line 12: | Line 12: | ||

[[File:equation1.png|700px]] | [[File:equation1.png|700px]] | ||

− | + | ||

+ | The goal of GANs is to train the generator g and the discriminator d and find a pair of (g,d) which the game reaches a Nash equilibrium. However, it could cause failure to converge since the generator and the discriminator is trained using gradient descent techniques which are not used to find a Nash equilibrium of a game. |

## Revision as of 17:06, 12 March 2018

## Introduction

Generative Adversarial Networks (GANs) are powerful generative models. It consists of a generator and a discriminator or critic. The generator is a neural network which learns to generate data as similar to the real distribution as possible. The critic measures the distance between the generated data distribution and the real data distribution which uses to distinguish between the generated data and the training data.

Optimal transport theory is a similar approach to measure the distance between the generated data and the training data distribution. The idea of this theory is by optimally transporting the data points between the training data and the generated data, the two data distribution will be approximately the same. This method produces a closed form solution instead of the approximated solution which is the advantage of this method. However, the primal form optimal transport could give biased gradients when mini-batches are applied.

This paper presents a variant GANs named OT-GAN, which incorporates a MIni-batch Energy Distance into its critic. This newly defined metric is unbiased when min-batch is applied.

## GANs AND OPTIMAL TRANSPORT

The objective function of the GAN:

The goal of GANs is to train the generator g and the discriminator d and find a pair of (g,d) which the game reaches a Nash equilibrium. However, it could cause failure to converge since the generator and the discriminator is trained using gradient descent techniques which are not used to find a Nash equilibrium of a game.