Energy-based Generative Adversarial Network

09/11/2016
by   Junbo Zhao, et al.
0

We introduce the "Energy-based Generative Adversarial Network" model (EBGAN) which views the discriminator as an energy function that attributes low energies to the regions near the data manifold and higher energies to other regions. Similar to the probabilistic GANs, a generator is seen as being trained to produce contrastive samples with minimal energies, while the discriminator is trained to assign high energies to these generated samples. Viewing the discriminator as an energy function allows to use a wide variety of architectures and loss functionals in addition to the usual binary classifier with logistic output. Among them, we show one instantiation of EBGAN framework as using an auto-encoder architecture, with the energy being the reconstruction error, in place of the discriminator. We show that this form of EBGAN exhibits more stable behavior than regular GANs during training. We also show that a single-scale architecture can be trained to generate high-resolution images.

READ FULL TEXT

page 7

page 8

page 9

page 15

page 16

page 17

research
11/16/2020

Mode Penalty Generative Adversarial Network with adapted Auto-encoder

Generative Adversarial Networks (GAN) are trained to generate sample ima...
research
03/05/2019

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

In this paper, we propose Orthogonal Generative Adversarial Networks (O-...
research
05/31/2017

Adversarial Ranking for Language Generation

Generative adversarial networks (GANs) have great successes on synthesiz...
research
07/17/2018

IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis

We present a novel introspective variational autoencoder (IntroVAE) mode...
research
09/06/2022

Improved Sensor-Based Animal Behavior Classification Performance through Conditional Generative Adversarial Network

Many activity classifications segments data into fixed window size for f...
research
04/02/2018

Updating the generator in PPGN-h with gradients flowing through the encoder

The Generative Adversarial Network framework has shown success in implic...
research
09/16/2019

Microstructure synthesis using style-based generative adversarial network

Work considers the usage of StyleGAN architecture for the task of micros...

Please sign up or login with your details

Forgot password? Click here to reset