DeepAI AI Chat
Log In Sign Up

Training Generative Adversarial Networks from Incomplete Observations using Factorised Discriminators

by   Daniel Stoller, et al.
Queen Mary University of London

Generative adversarial networks (GANs) have shown great success in applications such as image generation and inpainting. To stabilize the challenging training process, one typically requires large datasets - which are not available for many tasks. Large amounts of additionally available incomplete observations could be exploited in many cases, but it remains unclear how to train a GAN in such a setting. To address this shortcoming, we factorise the high-dimensional joint distribution of the complete data into a set of lower-dimensional distributions along with their dependencies. As a consequence, we can split the discriminator in a GAN into multiple "sub-discriminators" that can be independently trained from incomplete observations. Their outputs can be combined to obtain an estimate of the density ratio between the joint real and the generator distribution, which enables training the generator as in the original GAN framework. As an additional benefit, our modularisation facilitates incorporating prior knowledge into the discriminator architecture. We apply our method to image generation, image segmentation and audio source separation, and show an improved performance compared to a standard GAN when additional incomplete training examples are available.


page 16

page 17

page 18

page 19

page 20

page 21

page 23

page 24


Collaborative GAN Sampling

Generative adversarial networks (GANs) have shown great promise in gener...

MisGAN: Learning from Incomplete Data with Generative Adversarial Networks

Generative adversarial networks (GANs) have been shown to provide an eff...

GenCo: Generative Co-training on Data-Limited Image Generation

Training effective Generative Adversarial Networks (GANs) requires large...

Exploiting Pre-trained Feature Networks for Generative Adversarial Networks in Audio-domain Loop Generation

While generative adversarial networks (GANs) have been widely used in re...

AWGAN: Empowering High-Dimensional Discriminator Output for Generative Adversarial Networks

Empirically multidimensional discriminator (critic) output can be advant...

Generating Relevant Counter-Examples from a Positive Unlabeled Dataset for Image Classification

With surge of available but unlabeled data, Positive Unlabeled (PU) lear...

Blur, Noise, and Compression Robust Generative Adversarial Networks

Recently, generative adversarial networks (GANs), which learn data distr...