Movatterモバイル変換


[0]ホーム

URL:


loading
PapersPapers/2022PapersPapers/2022

Scitepress Logo

The Search is performed on all of the following fields:

Note: Please use complete words only.
  • Publication Title
  • Abstract
  • Publication Keywords
  • DOI
  • Proceeding Title
  • Proceeding Foreword
  • ISBN (Completed)
  • Insticc Ontology
  • Author Affiliation
  • Author Name
  • Editor Name
If you already have a Primoris Account you can use the same username/password here.
Research.Publish.Connect.

The Search is performed on all of the following fields:

Note: Please use complete words only.
  • Publication Title
  • Abstract
  • Publication Keywords
  • DOI
  • Proceeding Title
  • Proceeding Foreword
  • ISBN (Completed)
  • Insticc Ontology
  • Author Affiliation
  • Author Name
  • Editor Name
If you're looking for an exact phrase use quotation marks on text fields.

Paper

Paper Unlock

Authors:Jiaxu Zhao1;Li Huang1;Ruixuan Sun2;Liao Bing3 andHong Qu1

Affiliations:1University of Electronic Science and Technology of China, Chengdu 610054, China;2Yelp Inc., U.S.A.;3Chengdu Dajiangtong Technology Co., Ltd, China

Keyword(s):Neural Machine Translation, Exposure Bias, GAN.

Abstract:In recent years, Neural Machine Translation (NMT) has achieved great success, but we can not ignore two important problems. One is the exposure bias caused by the different strategies between training and inference, and the other is that the NMT model generates the best candidate word for the current step yet a bad element of the whole sentence. The popular methods to solve these two problems are Schedule Sampling and Generative Adversarial Networks (GANs) respectively, and both achieved some success. In this paper, we proposed a more precise approach called “similarity selection” combining a new GAN structure called twin-GAN to solve the above two problems. There are two generators and two discriminators in the twin-GAN. One generator uses the “similarity selection” and the other one uses the same way as inference (simulate the inference process). One discriminator guides generators at the sentence level, and the other discriminator forces these two generators to have similar distributions. Moreover, we performed a lot of experiments on the IWSLT 2014 German→English (De→En) and the WMT’17 Chinese!English (Zh→En) and the result shows that we improved the performance compared to some other strong baseline models which based on recurrentarchitecture.(More)

In recent years, Neural Machine Translation (NMT) has achieved great success, but we can not ignore two important problems. One is the exposure bias caused by the different strategies between training and inference, and the other is that the NMT model generates the best candidate word for the current step yet a bad element of the whole sentence. The popular methods to solve these two problems are Schedule Sampling and Generative Adversarial Networks (GANs) respectively, and both achieved some success. In this paper, we proposed a more precise approach called “similarity selection” combining a new GAN structure called twin-GAN to solve the above two problems. There are two generators and two discriminators in the twin-GAN. One generator uses the “similarity selection” and the other one uses the same way as inference (simulate the inference process). One discriminator guides generators at the sentence level, and the other discriminator forces these two generators to have similar distributions. Moreover, we performed a lot of experiments on the IWSLT 2014 German→English (De→En) and the WMT’17 Chinese!English (Zh→En) and the result shows that we improved the performance compared to some other strong baseline models which based on recurrent
architecture.

Full Text

Download
Please type the code

CC BY-NC-ND 4.0

Sign In

Guests can use SciTePress Digital Library without having a SciTePress account. However, guests have limited access to downloading full text versions of papers and no access to special options.
Guests can use SciTePress Digital Library without having a SciTePress account. However, guests have limited access to downloading full text versions of papers and no access to special options.
Guest:Register as new SciTePress user now for free.

Sign In

Download limit per month - 500 recent papers or 4000 papers more than 2 years old.
SciTePress user: please login.

PDF ImageMy Papers

PopUp Banner

Unable to see papers previously downloaded, because you haven't logged in as SciTePress Member.

If you are already a member please login.
You are not signed in, therefore limits apply to your IP address 153.126.140.213

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total
Popup Banner

PDF ButtonFull Text

Download
Please type the code

Paper citation in several formats:
Zhao, J., Huang, L., Sun, R., Bing, L. and Qu, H. (2021).Twin-GAN for Neural Machine Translation. InProceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART; ISBN 978-989-758-484-8; ISSN 2184-433X, SciTePress, pages 87-96. DOI: 10.5220/0010217300870096

@conference{icaart21,
author={Jiaxu Zhao and Li Huang and Ruixuan Sun and Liao Bing and Hong Qu},
title={Twin-GAN for Neural Machine Translation},
booktitle={Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART},
year={2021},
pages={87-96},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010217300870096},
isbn={978-989-758-484-8},
issn={2184-433X},
}

TY - CONF

JO - Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART
TI - Twin-GAN for Neural Machine Translation
SN - 978-989-758-484-8
IS - 2184-433X
AU - Zhao, J.
AU - Huang, L.
AU - Sun, R.
AU - Bing, L.
AU - Qu, H.
PY - 2021
SP - 87
EP - 96
DO - 10.5220/0010217300870096
PB - SciTePress

    - Science and Technology Publications, Lda.
    RESOURCES

    Proceedings

    Papers

    Authors

    Ontology

    CONTACTS

    Science and Technology Publications, Lda
    Avenida de S. Francisco Xavier, Lote 7 Cv. C,
    2900-616 Setúbal, Portugal.

    Phone: +351 265 520 185(National fixed network call)
    Fax: +351 265 520 186
    Email:info@scitepress.org

    EXTERNAL LINKS

    PRIMORIS

    INSTICC

    SCITEVENTS

    CROSSREF

    PROCEEDINGS SUBMITTED FOR INDEXATION BY:

    dblp

    Ei Compendex

    SCOPUS

    Semantic Scholar

    Google Scholar

    Microsoft Academic


    [8]
    ページ先頭

    ©2009-2025 Movatter.jp