Thesis Details
Neuronový strojový překlad pro jazykové páry s malým množstvím trénovacích dat
English title
Low-Resource Neural Machine Translation
Language
Czech
Abstract
This thesis deals with neural machine translation (NMT) for low-resource languages. The goal was to evaluate current techniques by using the experiments and suggest their improvements. The translation systems in this thesis used the neural network transformer architecture and were trained by the Marian framework. The selected language pairs were Slovak with Croatian and Slovak with Serbian. The subjects of the experiments were the transfer learning techniques and semi-supervised learning.
Keywords
neural machine translation, transformer, low-resource, transfer learning, semi-supervised learning, slavic languages, slovak, croatian
Department
Degree Programme
Information Technology
Files
Status
defended, grade C
Date
28 August 2020
Reviewer
Committee
Hliněná Dana, doc. RNDr., Ph.D. (DMAT FEEC BUT), předseda
Dytrych Jaroslav, Ing., Ph.D. (DCGM FIT BUT), člen
Herout Adam, prof. Ing., Ph.D. (DCGM FIT BUT), člen
Křena Bohuslav, Ing., Ph.D. (DITS FIT BUT), člen
Křivka Zbyněk, Ing., Ph.D. (DIFS FIT BUT), člen
Dytrych Jaroslav, Ing., Ph.D. (DCGM FIT BUT), člen
Herout Adam, prof. Ing., Ph.D. (DCGM FIT BUT), člen
Křena Bohuslav, Ing., Ph.D. (DITS FIT BUT), člen
Křivka Zbyněk, Ing., Ph.D. (DIFS FIT BUT), člen
Citation
FILO, Denis. Neuronový strojový překlad pro jazykové páry s malým množstvím trénovacích dat. Brno, 2020. Bachelor's Thesis. Brno University of Technology, Faculty of Information Technology. 2020-08-28. Supervised by Jon Josef. Available from: https://www.fit.vut.cz/study/thesis/23087/
BibTeX
@bachelorsthesis{FITBT23087, author = "Denis Filo", type = "Bachelor's thesis", title = "Neuronov\'{y} strojov\'{y} p\v{r}eklad pro jazykov\'{e} p\'{a}ry s mal\'{y}m mno\v{z}stv\'{i}m tr\'{e}novac\'{i}ch dat", school = "Brno University of Technology, Faculty of Information Technology", year = 2020, location = "Brno, CZ", language = "czech", url = "https://www.fit.vut.cz/study/thesis/23087/" }