@inproceedings{cai-etal-2022-advancing,
title = "Advancing Semi-Supervised Task Oriented Dialog Systems by {JSA} Learning of Discrete Latent Variable Models",
author = "Cai, Yucheng and
Liu, Hong and
Ou, Zhijian and
Huang, Yi and
Feng, Junlan",
editor = "Lemon, Oliver and
Hakkani-Tur, Dilek and
Li, Junyi Jessy and
Ashrafzadeh, Arash and
Garcia, Daniel Hern{\'a}ndez and
Alikhani, Malihe and
Vandyke, David and
Du{\v{s}}ek, Ond{\v{r}}ej",
booktitle = "Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue",
month = sep,
year = "2022",
address = "Edinburgh, UK",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.sigdial-1.44",
doi = "10.18653/v1/2022.sigdial-1.44",
pages = "456--467",
abstract = "Developing semi-supervised task-oriented dialog (TOD) systems by leveraging unlabeled dialog data has attracted increasing interests. For semi-supervised learning of latent state TOD models, variational learning is often used, but suffers from the annoying high-variance of the gradients propagated through discrete latent variables and the drawback of indirectly optimizing the target log-likelihood. Recently, an alternative algorithm, called joint stochastic approximation (JSA), has emerged for learning discrete latent variable models with impressive performances. In this paper, we propose to apply JSA to semi-supervised learning of the latent state TOD models, which is referred to as JSA-TOD. To our knowledge, JSA-TOD represents the first work in developing JSA based semi-supervised learning of discrete latent variable conditional models for such long sequential generation problems like in TOD systems. Extensive experiments show that JSA-TOD significantly outperforms its variational learning counterpart. Remarkably, semi-supervised JSA-TOD using 20{\%} labels performs close to the full-supervised baseline on MultiWOZ2.1.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cai-etal-2022-advancing">
<titleInfo>
<title>Advancing Semi-Supervised Task Oriented Dialog Systems by JSA Learning of Discrete Latent Variable Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yucheng</namePart>
<namePart type="family">Cai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hong</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhijian</namePart>
<namePart type="family">Ou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junlan</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Oliver</namePart>
<namePart type="family">Lemon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dilek</namePart>
<namePart type="family">Hakkani-Tur</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junyi</namePart>
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arash</namePart>
<namePart type="family">Ashrafzadeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="given">Hernández</namePart>
<namePart type="family">Garcia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Malihe</namePart>
<namePart type="family">Alikhani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Vandyke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ondřej</namePart>
<namePart type="family">Dušek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Edinburgh, UK</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Developing semi-supervised task-oriented dialog (TOD) systems by leveraging unlabeled dialog data has attracted increasing interests. For semi-supervised learning of latent state TOD models, variational learning is often used, but suffers from the annoying high-variance of the gradients propagated through discrete latent variables and the drawback of indirectly optimizing the target log-likelihood. Recently, an alternative algorithm, called joint stochastic approximation (JSA), has emerged for learning discrete latent variable models with impressive performances. In this paper, we propose to apply JSA to semi-supervised learning of the latent state TOD models, which is referred to as JSA-TOD. To our knowledge, JSA-TOD represents the first work in developing JSA based semi-supervised learning of discrete latent variable conditional models for such long sequential generation problems like in TOD systems. Extensive experiments show that JSA-TOD significantly outperforms its variational learning counterpart. Remarkably, semi-supervised JSA-TOD using 20% labels performs close to the full-supervised baseline on MultiWOZ2.1.</abstract>
<identifier type="citekey">cai-etal-2022-advancing</identifier>
<identifier type="doi">10.18653/v1/2022.sigdial-1.44</identifier>
<location>
<url>https://aclanthology.org/2022.sigdial-1.44</url>
</location>
<part>
<date>2022-09</date>
<extent unit="page">
<start>456</start>
<end>467</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Advancing Semi-Supervised Task Oriented Dialog Systems by JSA Learning of Discrete Latent Variable Models
%A Cai, Yucheng
%A Liu, Hong
%A Ou, Zhijian
%A Huang, Yi
%A Feng, Junlan
%Y Lemon, Oliver
%Y Hakkani-Tur, Dilek
%Y Li, Junyi Jessy
%Y Ashrafzadeh, Arash
%Y Garcia, Daniel Hernández
%Y Alikhani, Malihe
%Y Vandyke, David
%Y Dušek, Ondřej
%S Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue
%D 2022
%8 September
%I Association for Computational Linguistics
%C Edinburgh, UK
%F cai-etal-2022-advancing
%X Developing semi-supervised task-oriented dialog (TOD) systems by leveraging unlabeled dialog data has attracted increasing interests. For semi-supervised learning of latent state TOD models, variational learning is often used, but suffers from the annoying high-variance of the gradients propagated through discrete latent variables and the drawback of indirectly optimizing the target log-likelihood. Recently, an alternative algorithm, called joint stochastic approximation (JSA), has emerged for learning discrete latent variable models with impressive performances. In this paper, we propose to apply JSA to semi-supervised learning of the latent state TOD models, which is referred to as JSA-TOD. To our knowledge, JSA-TOD represents the first work in developing JSA based semi-supervised learning of discrete latent variable conditional models for such long sequential generation problems like in TOD systems. Extensive experiments show that JSA-TOD significantly outperforms its variational learning counterpart. Remarkably, semi-supervised JSA-TOD using 20% labels performs close to the full-supervised baseline on MultiWOZ2.1.
%R 10.18653/v1/2022.sigdial-1.44
%U https://aclanthology.org/2022.sigdial-1.44
%U https://doi.org/10.18653/v1/2022.sigdial-1.44
%P 456-467
Markdown (Informal)
[Advancing Semi-Supervised Task Oriented Dialog Systems by JSA Learning of Discrete Latent Variable Models](https://aclanthology.org/2022.sigdial-1.44) (Cai et al., SIGDIAL 2022)
ACL