Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Quasi-Symplectic Langevin Variational Autoencoder

Zihao Wang 1, * Hervé Delingette 1
* Corresponding author
Abstract : Variational autoencoder (VAE) is a very popular and well-investigated generative model vastly used in neural learning research. To leverage VAE in practical tasks dealing with a massive dataset of large dimensions it is required to deal with the difficulty of building low variance evidence lower bounds (ELBO). Markov Chain Monte Carlo (MCMC) is one of the effective approaches to tighten the ELBO for approximating the posterior distribution. Hamiltonian Variational Autoencoder (HVAE) is an effective MCMC inspired approach for constructing a the lowvariance ELBO which is also amenable to the reparameterization trick. In this work, we propose a Quasi-symplectic Langevin Variational autoencoder (LangevinVAE) by incorporating the gradients information in the inference process through Langevin dynamic. We shows the effectiveness of the proposed approach by toy and real world examples.
Document type :
Preprints, Working Papers, ...
Complete list of metadata

https://hal.inria.fr/hal-03024748
Contributor : Zihao Wang <>
Submitted on : Monday, April 19, 2021 - 4:16:03 PM
Last modification on : Tuesday, April 20, 2021 - 3:29:11 AM

File

Langevin_VAE_Revision.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03024748, version 3

Collections

Citation

Zihao Wang, Hervé Delingette. Quasi-Symplectic Langevin Variational Autoencoder. 2021. ⟨hal-03024748v3⟩

Share

Metrics

Record views

12

Files downloads

96