O-GAN: Extremely Concise Approach for Auto-Encoding Generative Adversarial Networks

03/05/2019
by   Jianlin Su, et al.
0

In this paper, we propose Orthogonal Generative Adversarial Networks (O-GANs). We decompose the network of discriminator orthogonally and add an extra loss into the objective of common GANs, which can enforce discriminator become an effective encoder. The same extra loss can be embedded into any kind of GANs and there is almost no increase in computation. Furthermore, we discuss the principle of our method, which is relative to the fully-exploiting of the remaining degrees of freedom of discriminator. As we know, our solution is the simplest approach to train a generative adversarial network with auto-encoding ability.

READ FULL TEXT

page 6

page 7

research
06/05/2016

Semi-Supervised Learning with Generative Adversarial Networks

We extend Generative Adversarial Networks (GANs) to the semi-supervised ...
research
06/15/2017

Variational Approaches for Auto-Encoding Generative Adversarial Networks

Auto-encoding generative adversarial networks (GANs) combine the standar...
research
03/24/2018

MTGAN: Speaker Verification through Multitasking Triplet Generative Adversarial Networks

In this paper, we propose an enhanced triplet method that improves the e...
research
09/11/2016

Energy-based Generative Adversarial Network

We introduce the "Energy-based Generative Adversarial Network" model (EB...
research
06/01/2018

Unsupervised Object Localization using Generative Adversarial Networks

This paper introduces a novel end-to-end deep neural network model for u...
research
12/19/2018

RankGAN: A Maximum Margin Ranking GAN for Generating Faces

We present a new stage-wise learning paradigm for training generative ad...
research
09/05/2017

Linking Generative Adversarial Learning and Binary Classification

In this note, we point out a basic link between generative adversarial (...

Please sign up or login with your details

Forgot password? Click here to reset