Data-Dependent Conditional Priors for Unsupervised Learning of Multimodal Data

  • One of the major shortcomings of variational autoencoders is the inability to produce generations from the individual modalities of data originating from mixture distributions. This is primarily due to the use of a simple isotropic Gaussian as the prior for the latent code in the ancestral sampling procedure for data generations. In this paper, we propose a novel formulation of variational autoencoders, conditional prior VAE (CP-VAE), with a two-level generative process for the observed data where continuous 𝐳 and a discrete 𝐜 variables are introduced in addition to the observed variables 𝐱. By learning data-dependent conditional priors, the new variational objective naturally encourages a better match between the posterior and prior conditionals, and the learning of the latent categories encoding the major source of variation of the original data in an unsupervised manner. Through sampling continuous latent code from the data-dependent conditional priors, we are able to generate new samples from the individual mixture components corresponding, to the multimodal structure over the original data. Moreover, we unify and analyse our objective under different independence assumptions for the joint distribution of the continuous and discrete latent variables. We provide an empirical evaluation on one synthetic dataset and three image datasets, FashionMNIST, MNIST, and Omniglot, illustrating the generative performance of our new model comparing to multiple baselines.

Export metadata

Additional Services

Search Google Scholar
Metadaten
Author:Frantzeska Lavda, Magda Gregorová, Alexandros Kalousis
DOI:https://doi.org/10.3390/e22080888
Parent Title (English):Entropy
Document Type:Article
Language:English
Year of publication:2020
Release Date:2024/01/22
Volume:22
Issue:8
Pages/Size:1
First Page:888
Last Page:888
Verstanden ✔
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.