| layout | img | img_link | title | img2 | img2_link | active_tab |
|---|---|---|---|---|---|---|
default |
cover |
Machine Translation |
nmt-book.jpg |
main_page |
EN 601.468/668 Machine Translation : Fall 2024 : Tuesdays and Thursdays 1:30-2:45 : Room: Hodson 210 : Computer Science Department : Johns Hopkins University
Google translate instantly translates between any pair of over eighty human languages like French and English. How does it do that? Why does it make the errors that it does? And how can you build something better? Modern translation systems like Google Translate and Bing Translator learn to translate by reading millions of words of already translated text. This course will show you how they work. We cover fundamental building blocks from linguistics, machine learning (especially deep learning), algorithms, and data structures, showing how they apply to a difficult real-word artificial intelligence problem.
Instructor : Philipp Koehn ([email protected])
TA : Bismarck Odoom ([email protected])
Office hours : Professor by Appointment : TA Bismarck Bamfo Odoom, Wednesday 1-2pm, Malone 122 : CAs Lavanya Shankar, Monday 3-4pm; Angad Sandhu, Tuesday 1-2pm; Kshitij Joshi, Friday 2-3pm; Weina Dai, Thursday 2-3pm; Malone 216
Format : This class will be delivered in person this year. Lectures were pre-recorded in 2020 and these are still accessible through links on this web site and on Youtube. Note that a good portion has changed since.
Discussion Forum : Piazza access code: nobxprze2nm ! Gradescope access code: 7EXJ44
Textbooks : The class follows closely two textbooks.
- Statistical Machine Translation (errata) by Philipp Koehn, 2010 (JHU library or Amazon).
- Neural Machine Translation (errata), by Philipp Koehn, 2020 (JHU library or Amazon).
Grading : To understand how machine translation works, you will build a translation system. We will mainly grade hands-on work.
- Five homework assignments (12% each)
- Final project (30%)
- In-class presentation: Language in ten minutes (10%)
Homework Schedule : There will be five homework assignments, tentative schedule:
- HW1: Analysis, due September 5
- HW2: Word alignment, due September 19
- HW3: Decoding, due October 3
- HW4: Neural translation model part 1, due October 17
- HW5: Neural translation model part 2, due October 31
Late penalty for homework assignments: 10% per day, after 5 free "late days".