@inproceedings{song-etal-2024-scale,
title = "Scale-{VAE}: Preventing Posterior Collapse in Variational Autoencoder",
author = "Song, Tianbao and
Sun, Jingbo and
Liu, Xin and
Peng, Weiming",
editor = "Calzolari, Nicoletta and
Kan, Min-Yen and
Hoste, Veronique and
Lenci, Alessandro and
Sakti, Sakriani and
Xue, Nianwen",
booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
month = may,
year = "2024",
address = "Torino, Italia",
publisher = "ELRA and ICCL",
url = "https://aclanthology.org/2024.lrec-main.1250",
pages = "14347--14357",
abstract = "Variational autoencoder (VAE) is a widely used generative model that gains great popularity for its capability in density estimation and representation learning. However, when employing a strong autoregressive generation network, VAE tends to converge to a degenerate local optimum known as posterior collapse. In this paper, we propose a model named Scale-VAE to solve this problem. Scale-VAE does not force the KL term to be larger than a positive constant, but aims to make the latent variables easier to be exploited by the generation network. Specifically, each dimension of the mean for the approximate posterior distribution is multiplied by a factor to keep that dimension discriminative across data instances. The same factors are used for all data instances so as not to change the relative relationship between the posterior distributions. Latent variables from the scaled-up posteriors are fed into the generation network, but the original posteriors are still used to calculate the KL term. In this way, Scale-VAE can solve the posterior collapse problem with a training cost similar to or even lower than the basic VAE. Experimental results show that Scale-VAE outperforms state-of-the-art models in density estimation, representation learning, and consistency of the latent space, and is competitive with other models in generation.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="song-etal-2024-scale">
<titleInfo>
<title>Scale-VAE: Preventing Posterior Collapse in Variational Autoencoder</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tianbao</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingbo</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weiming</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicoletta</namePart>
<namePart type="family">Calzolari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min-Yen</namePart>
<namePart type="family">Kan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Veronique</namePart>
<namePart type="family">Hoste</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Lenci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nianwen</namePart>
<namePart type="family">Xue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA and ICCL</publisher>
<place>
<placeTerm type="text">Torino, Italia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Variational autoencoder (VAE) is a widely used generative model that gains great popularity for its capability in density estimation and representation learning. However, when employing a strong autoregressive generation network, VAE tends to converge to a degenerate local optimum known as posterior collapse. In this paper, we propose a model named Scale-VAE to solve this problem. Scale-VAE does not force the KL term to be larger than a positive constant, but aims to make the latent variables easier to be exploited by the generation network. Specifically, each dimension of the mean for the approximate posterior distribution is multiplied by a factor to keep that dimension discriminative across data instances. The same factors are used for all data instances so as not to change the relative relationship between the posterior distributions. Latent variables from the scaled-up posteriors are fed into the generation network, but the original posteriors are still used to calculate the KL term. In this way, Scale-VAE can solve the posterior collapse problem with a training cost similar to or even lower than the basic VAE. Experimental results show that Scale-VAE outperforms state-of-the-art models in density estimation, representation learning, and consistency of the latent space, and is competitive with other models in generation.</abstract>
<identifier type="citekey">song-etal-2024-scale</identifier>
<location>
<url>https://aclanthology.org/2024.lrec-main.1250</url>
</location>
<part>
<date>2024-05</date>
<extent unit="page">
<start>14347</start>
<end>14357</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Scale-VAE: Preventing Posterior Collapse in Variational Autoencoder
%A Song, Tianbao
%A Sun, Jingbo
%A Liu, Xin
%A Peng, Weiming
%Y Calzolari, Nicoletta
%Y Kan, Min-Yen
%Y Hoste, Veronique
%Y Lenci, Alessandro
%Y Sakti, Sakriani
%Y Xue, Nianwen
%S Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
%D 2024
%8 May
%I ELRA and ICCL
%C Torino, Italia
%F song-etal-2024-scale
%X Variational autoencoder (VAE) is a widely used generative model that gains great popularity for its capability in density estimation and representation learning. However, when employing a strong autoregressive generation network, VAE tends to converge to a degenerate local optimum known as posterior collapse. In this paper, we propose a model named Scale-VAE to solve this problem. Scale-VAE does not force the KL term to be larger than a positive constant, but aims to make the latent variables easier to be exploited by the generation network. Specifically, each dimension of the mean for the approximate posterior distribution is multiplied by a factor to keep that dimension discriminative across data instances. The same factors are used for all data instances so as not to change the relative relationship between the posterior distributions. Latent variables from the scaled-up posteriors are fed into the generation network, but the original posteriors are still used to calculate the KL term. In this way, Scale-VAE can solve the posterior collapse problem with a training cost similar to or even lower than the basic VAE. Experimental results show that Scale-VAE outperforms state-of-the-art models in density estimation, representation learning, and consistency of the latent space, and is competitive with other models in generation.
%U https://aclanthology.org/2024.lrec-main.1250
%P 14347-14357
Markdown (Informal)
[Scale-VAE: Preventing Posterior Collapse in Variational Autoencoder](https://aclanthology.org/2024.lrec-main.1250) (Song et al., LREC-COLING 2024)
ACL