Learning coupled policies for simultaneous machine translation using imitation learning

Philip Arthur, Trevor Cohn, Gholamreza Haffari

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

Abstract

We present a novel approach to efficiently learn a simultaneous translation model with coupled programmer-interpreter policies. First, we present an algorithmic oracle to produce oracle READ/WRITE actions for training bilingual sentence-pairs using the notion of word alignments. This oracle actions are designed to capture enough information from the partial input before writing the output. Next, we perform a coupled scheduled sampling to effectively mitigate the exposure bias when learning both policies jointly with imitation learning. Experiments on six language-pairs show our method outperforms strong baselines in terms of translation quality while keeping the translation delay low.

Original languageEnglish
Title of host publicationThe 16th Conference of the European Chapter of the Association for Computational Linguistics
EditorsValerio Basile, Tommaso Caselli
Place of PublicationStroudsburg PA USA
PublisherAssociation for Computational Linguistics (ACL)
Pages2709-2719
Number of pages11
ISBN (Electronic)9781954085022
Publication statusPublished - 2021
EventEuropean Association of Computational Linguistics Conference 2021 - Virtual, Virtual, Online, United States of America
Duration: 19 Apr 202123 Apr 2021
Conference number: 16th
https://www.aclweb.org/anthology/volumes/2021.eacl-main/ (Proceedings)
https://2021.eacl.org/ (Website)

Conference

ConferenceEuropean Association of Computational Linguistics Conference 2021
Abbreviated titleEACL 2021
Country/TerritoryUnited States of America
CityVirtual, Online
Period19/04/2123/04/21
Internet address

Cite this