DeepAI AI Chat
Log In Sign Up

When do GANs replicate? On the choice of dataset size

by   Qianli Feng, et al.
The Ohio State University

Do GANs replicate training images? Previous studies have shown that GANs do not seem to replicate training data without significant change in the training procedure. This leads to a series of research on the exact condition needed for GANs to overfit to the training data. Although a number of factors has been theoretically or empirically identified, the effect of dataset size and complexity on GANs replication is still unknown. With empirical evidence from BigGAN and StyleGAN2, on datasets CelebA, Flower and LSUN-bedroom, we show that dataset size and its complexity play an important role in GANs replication and perceptual quality of the generated images. We further quantify this relationship, discovering that replication percentage decays exponentially with respect to dataset size and complexity, with a shared decaying factor across GAN-dataset combinations. Meanwhile, the perceptual image quality follows a U-shape trend w.r.t dataset size. This finding leads to a practical tool for one-shot estimation on minimal dataset size to prevent GAN replication which can be used to guide datasets construction and selection.


page 1

page 4


Visual Indeterminacy in Generative Neural Art

Why are GANs such powerful tools for making art? This essay argues that ...

Image Augmentations for GAN Training

Data augmentations have been widely studied to improve the accuracy and ...

Image Generation and Recognition (Emotions)

Generative Adversarial Networks (GANs) were proposed in 2014 by Goodfell...

Empirical Analysis of Overfitting and Mode Drop in GAN Training

We examine two key questions in GAN training, namely overfitting and mod...

Financial factors selection with knockoffs: fund replication, explanatory and prediction networks

We apply the knockoff procedure to factor selection in finance. By build...

Identifying Statistical Bias in Dataset Replication

Dataset replication is a useful tool for assessing whether improvements ...

Generating Symbolic Reasoning Problems with Transformer GANs

Constructing training data for symbolic reasoning domains is challenging...