Implications of data topology for deep generative models

Implications of data topology for deep generative models

Many deep generative models, such as variational autoencoders (VAEs) and generative adversarial networks (GANs), learn an immersion mapping from a standard normal distribution in a low-dimensional latent space into a higher-dimensional data space. As such, these mappings are only capable of producin...

Full description

Bibliographic Details
Main Authors: Yinzhu Jin, Rory McDaniel, N. Joseph Tatro, Michael J. Catanzaro, Abraham D. Smith, Paul Bendich, Matthew B. Dwyer, P. Thomas Fletcher
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-08-01
Series:Frontiers in Computer Science
Subjects:
data topology
generative model
variational autoencoder (VAE)
diffusion probabilistic models (DDPM)
topological data analysis
Online Access:https://www.frontiersin.org/articles/10.3389/fcomp.2024.1260604/full

Internet

https://www.frontiersin.org/articles/10.3389/fcomp.2024.1260604/full

Similar Items