Latent Variable Modeling for Generative Concept Representations and Deep Generative Models
Machine Learning
2019-01-01 v1 Machine Learning
Abstract
Latent representations are the essence of deep generative models and determine their usefulness and power. For latent representations to be useful as generative concept representations, their latent space must support latent space interpolation, attribute vectors and concept vectors, among other things. We investigate and discuss latent variable modeling, including latent variable models, latent representations and latent spaces, particularly hierarchical latent representations and latent space vectors and geometry. Our focus is on that used in variational autoencoders and generative adversarial networks.
Cite
@article{arxiv.1812.11856,
title = {Latent Variable Modeling for Generative Concept Representations and Deep Generative Models},
author = {Daniel T. Chang},
journal= {arXiv preprint arXiv:1812.11856},
year = {2019}
}
Comments
arXiv admin note: text overlap with arXiv:1706.00400 by other authors