Escaping from Collapsing Modes in a Constrained Space

08/22/2018
by   Chia-Che Chang, et al.
0

Generative adversarial networks (GANs) often suffer from unpredictable mode-collapsing during training. We study the issue of mode collapse of Boundary Equilibrium Generative Adversarial Network (BEGAN), which is one of the state-of-the-art generative models. Despite its potential of generating high-quality images, we find that BEGAN tends to collapse at some modes after a period of training. We propose a new model, called BEGAN with a Constrained Space (BEGAN-CS), which includes a latent-space constraint in the loss function. We show that BEGAN-CS can significantly improve training stability and suppress mode collapse without either increasing the model complexity or degrading the image quality. Further, we visualize the distribution of latent vectors to elucidate the effect of latent-space constraint. The experimental results show that our method has additional advantages of being able to train on small datasets and to generate images similar to a given real image yet with variations of designated attributes on-the-fly.

READ FULL TEXT
research
04/11/2020

Autoencoding Generative Adversarial Networks

In the years since Goodfellow et al. introduced Generative Adversarial N...
research
08/19/2020

Improving Text to Image Generation using Mode-seeking Function

Generative Adversarial Networks (GANs) have long been used to understand...
research
11/25/2020

Multiclass non-Adversarial Image Synthesis, with Application to Classification from Very Small Sample

The generation of synthetic images is currently being dominated by Gener...
research
10/28/2022

Latent Space is Feature Space: Regularization Term for GANs Training on Limited Dataset

Generative Adversarial Networks (GAN) is currently widely used as an uns...
research
03/19/2023

Elastic Interaction Energy-Based Generative Model: Approximation in Feature Space

In this paper, we propose a novel approach to generative modeling using ...
research
08/06/2017

Image Quality Assessment Techniques Show Improved Training and Evaluation of Autoencoder Generative Adversarial Networks

We propose a training and evaluation approach for autoencoder Generative...
research
07/13/2019

Neural Embedding for Physical Manipulations

In common real-world robotic operations, action and state spaces can be ...

Please sign up or login with your details

Forgot password? Click here to reset