CUNI NMT System for WAT 2018 Translation Tasks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F18%3A10390164" target="_blank" >RIV/00216208:11320/18:10390164 - isvavai.cz</a>

Výsledek na webu

<a href="http://lotus.kuee.kyoto-u.ac.jp/WAT/WAT2018/WAT2018-proceedings-20181204.zip" target="_blank" >http://lotus.kuee.kyoto-u.ac.jp/WAT/WAT2018/WAT2018-proceedings-20181204.zip</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

CUNI NMT System for WAT 2018 Translation Tasks

Popis výsledku v původním jazyce

This paper describes the CUNI submission to WAT 2018 for the English-Hindi translation task using a transfer learning techniques which has proven effective under low resource conditions. We have used the Transformer model and utilized an English-Czech parallel corpus as additional data source. Our simple transfer learning approach first trains a &quot;parent&quot; model for a high-resource language pair (English-Czech) and then continues the training on the low-resource (English-Hindi) pair by replacing the training corpus. This setup improves the performance compared with the baseline and in combination with back-translation of Hindi monolingual data, it allowed us to win the English-Hindi task. The automatic scoring by BLEU did not correlate well with human judgments.

Název v anglickém jazyce

CUNI NMT System for WAT 2018 Translation Tasks

Popis výsledku anglicky

This paper describes the CUNI submission to WAT 2018 for the English-Hindi translation task using a transfer learning techniques which has proven effective under low resource conditions. We have used the Transformer model and utilized an English-Czech parallel corpus as additional data source. Our simple transfer learning approach first trains a &quot;parent&quot; model for a high-resource language pair (English-Czech) and then continues the training on the low-resource (English-Hindi) pair by replacing the training corpus. This setup improves the performance compared with the baseline and in combination with back-translation of Hindi monolingual data, it allowed us to win the English-Hindi task. The automatic scoring by BLEU did not correlate well with human judgments.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

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

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

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

Rok uplatnění

2018

Kód důvěrnosti údajů

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