U kunt zich hieronder aanmelden voor onze nieuwsbrief, zodat u op de hoogte wordt gehouden…
|Under the hood of Neural MT: a non-technical tour
Vincent Vandeghinste, KU Leuven
This talk will provide an insight into the inner workings of the new paradigm of neural machine translation (NMT), compared to the older paradigm of statistical machine translation. NMT will be explained without requiring a mathematical or technical background. We will highlight the strengths and weaknesses of NMT, so you will know what to expect from these systems.
Doelgroep: vertalers, gebruikers van automatische vertaling
Dr. Vincent Vandeghinste is a post-doctoral researcher at the Instituut voor de Nederlandse Taal and at the KU Leuven, and has been working on natural language processing and translation technologies since 1998. He was the project coordinator of the SCATE project (Smart Computer-Aided Translation Environment), a Flemish four year research project about improving the translation environment and (co)-authored about 70 publications in the areas of corpus building, treebanking, machine translation, augmented alternative communication and text-to-pictograph translation. He teaches Natural Language Processing, Language Engineering Applications and Linguistics And Artificial Intelligence in the advanced masters program for Artificial Intelligence at KU Leuven, as well as Computational Linguistics to students of Linguistics and Translation students. He has been working on machine translation since 2004.