A hybrid convolutional variational autoencoder for text generation

Stanislau Semeniuta, Aliaksei Severyn, Erhardt Barth


In this paper we explore the effect of architectural choices on learning a variational autoencoder (VAE) for text generation. In contrast to the previously introduced VAE model for text where both the encoder and decoder are RNNs, we propose a novel hybrid architecture that blends fully feed-forward convolutional and deconvolutional components with a recurrent language model. Our architecture exhibits several attractive properties such as faster run time and convergence, ability to better handle long sequences and, more importantly, it helps to avoid the issue of the VAE collapsing to a deterministic model.

Original languageEnglish
Title of host publicationEMNLP 2017 - Conference on Empirical Methods in Natural Language Processing, Proceedings
Number of pages11
PublisherAssociation for Computational Linguistics (ACL)
Publication date09.2017
ISBN (Print)978-194562683-8
Publication statusPublished - 09.2017
Event2017 Conference on Empirical Methods in Natural Language Processing - Copenhagen, Denmark
Duration: 09.09.201711.09.2017
Conference number: 150071

Cite this