Transfer learning and subword sampling for asymmetric-resource one-to-many neural translation

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10426959" target="_blank" >RIV/00216208:11320/20:10426959 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/s10590-020-09253-x" target="_blank" >https://doi.org/10.1007/s10590-020-09253-x</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Transfer learning and subword sampling for asymmetric-resource one-to-many neural translation

Popis výsledku v původním jazyce

There are several approaches for improving neural machine translation for low-resource languages: monolingual data can be exploited via pretraining or data augmentation; parallel corpora on related language pairs can be used via parameter sharing or transfer learning in multilingual models; subword segmentation and regularization techniques can be applied to ensure high coverage of the vocabulary. We review these approaches in the context of an asymmetric-resource one-to-many translation task, in which the pair of target languages are related, with one being a very low-resource and the other a higher-resource language. We test various methods on three artificially restricted translation tasks—English to Estonian (low-resource) and Finnish (high-resource), English to Slovak and Czech, English to Danish and Swedish—and one real-world task, Norwegian to North Sámi and Finnish. The experiments show positive effects especially for scheduled multi-task learning, denoising autoencoder, and subword sampling.

Název v anglickém jazyce

Transfer learning and subword sampling for asymmetric-resource one-to-many neural translation

Popis výsledku anglicky

There are several approaches for improving neural machine translation for low-resource languages: monolingual data can be exploited via pretraining or data augmentation; parallel corpora on related language pairs can be used via parameter sharing or transfer learning in multilingual models; subword segmentation and regularization techniques can be applied to ensure high coverage of the vocabulary. We review these approaches in the context of an asymmetric-resource one-to-many translation task, in which the pair of target languages are related, with one being a very low-resource and the other a higher-resource language. We test various methods on three artificially restricted translation tasks—English to Estonian (low-resource) and Finnish (high-resource), English to Slovak and Czech, English to Danish and Swedish—and one real-world task, Norwegian to North Sámi and Finnish. The experiments show positive effects especially for scheduled multi-task learning, denoising autoencoder, and subword sampling.

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

—

Návaznosti

—

Rok uplatnění

2020

Kód důvěrnosti údajů

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