Unsupervised Pretraining for Neural Machine Translation Using Elastic Weight Consolidation

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F19%3A10405613" target="_blank" >RIV/00216208:11320/19:10405613 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.aclweb.org/anthology/P19-2017.pdf" target="_blank" >https://www.aclweb.org/anthology/P19-2017.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Unsupervised Pretraining for Neural Machine Translation Using Elastic Weight Consolidation

Popis výsledku v původním jazyce

This work presents our ongoing research of unsupervised pretraining in neural machine translation (NMT). In our method, we initialize the weights of the encoder and decoder with two language models that are trained with monolingual data and then fine-tune the model on parallel data using Elastic Weight Consolidation (EWC) to avoid forgetting of the original language modeling tasks. We compare the regularization by EWC with the previous work that focuses on regularization by language modeling objectives. The positive result is that using EWC with the decoder achieves BLEU scores similar to the previous work. However, the model converges 2-3 times faster and does not require the original unlabeled training data during the finetuning stage. In contrast, the regularization using EWC is less effective if the original and new tasks are not closely related. We show that initializing the bidirectional NMT encoder with a left-toright language model and forcing the model to remember the original left-to-right l

Název v anglickém jazyce

Unsupervised Pretraining for Neural Machine Translation Using Elastic Weight Consolidation

Popis výsledku anglicky

This work presents our ongoing research of unsupervised pretraining in neural machine translation (NMT). In our method, we initialize the weights of the encoder and decoder with two language models that are trained with monolingual data and then fine-tune the model on parallel data using Elastic Weight Consolidation (EWC) to avoid forgetting of the original language modeling tasks. We compare the regularization by EWC with the previous work that focuses on regularization by language modeling objectives. The positive result is that using EWC with the decoder achieves BLEU scores similar to the previous work. However, the model converges 2-3 times faster and does not require the original unlabeled training data during the finetuning stage. In contrast, the regularization using EWC is less effective if the original and new tasks are not closely related. We show that initializing the bidirectional NMT encoder with a left-toright language model and forcing the model to remember the original left-to-right l

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GX19-26934X" target="_blank" >GX19-26934X: Neuronové reprezentace v multimodálním a mnohojazyčném modelování</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2019

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop

ISBN

978-1-950737-47-5

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

130-135

Název nakladatele

Association for Computational Linguistics

Místo vydání

Stroudsburg, PA, USA

Místo konání akce

Firenze, Italy

Datum konání akce

28. 7. 2019

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—