English

Dynamical Variational Autoencoders: A Comprehensive Review

Machine Learning 2022-07-05 v4 Machine Learning

Abstract

Variational autoencoders (VAEs) are powerful deep generative models widely used to represent high-dimensional complex data through a low-dimensional latent space learned in an unsupervised manner. In the original VAE model, the input data vectors are processed independently. Recently, a series of papers have presented different extensions of the VAE to process sequential data, which model not only the latent space but also the temporal dependencies within a sequence of data vectors and corresponding latent vectors, relying on recurrent neural networks or state-space models. In this paper, we perform a literature review of these models. We introduce and discuss a general class of models, called dynamical variational autoencoders (DVAEs), which encompasses a large subset of these temporal VAE extensions. Then, we present in detail seven recently proposed DVAE models, with an aim to homogenize the notations and presentation lines, as well as to relate these models with existing classical temporal models. We have reimplemented those seven DVAE models and present the results of an experimental benchmark conducted on the speech analysis-resynthesis task (the PyTorch code is made publicly available). The paper concludes with a discussion on important issues concerning the DVAE class of models and future research guidelines.

Keywords

Cite

@article{arxiv.2008.12595,
  title  = {Dynamical Variational Autoencoders: A Comprehensive Review},
  author = {Laurent Girin and Simon Leglaive and Xiaoyu Bie and Julien Diard and Thomas Hueber and Xavier Alameda-Pineda},
  journal= {arXiv preprint arXiv:2008.12595},
  year   = {2022}
}
R2 v1 2026-06-23T18:09:47.566Z