Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
MOSCOW +7(495)-142-86-81

Overview of deep learning methods for neural machine translation

Chzhun Zhujyuj   (Postgraduate student, National Research University Moscow Institute of Electronic Technology)

The article discusses the main methods of neurolinguistic training. Three stages of machine translation development are revealed: rule-based, statistical, neural learning. Two popular methods of neural machine translation are described: recurrent and convolutional. Advanced neural machine learning methods are characterized: ConvS2S, Transformer, RNMT +. A forecast was made for the further development of neural text translation systems.

Keywords:machine translation, neural learning, neural learning methods, neurolinguistics, ConvS2S, Transformer, RNMT +.

 

Read the full article …



Citation link:
Chzhun Z. Overview of deep learning methods for neural machine translation // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2021. -№03. -С. 223-230 DOI 10.37882/2223-2966.2021.03.36
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"