Generalized Latent Variable Recovery for Generative Adversarial Networks
Machine Learning
2018-10-10 v1 Machine Learning
Abstract
The Generator of a Generative Adversarial Network (GAN) is trained to transform latent vectors drawn from a prior distribution into realistic looking photos. These latent vectors have been shown to encode information about the content of their corresponding images. Projecting input images onto the latent space of a GAN is non-trivial, but previous work has successfully performed this task for latent spaces with a uniform prior. We extend these techniques to latent spaces with a Gaussian prior, and demonstrate our technique's effectiveness.
Cite
@article{arxiv.1810.03764,
title = {Generalized Latent Variable Recovery for Generative Adversarial Networks},
author = {Nicholas Egan and Jeffrey Zhang and Kevin Shen},
journal= {arXiv preprint arXiv:1810.03764},
year = {2018}
}