CSE 521 - Machine Translation

The course aims to introduce students to the field of machine translation. The main objective of the course is to show you how modern translation systems learn to translate by reading millions of words. The goal is to provide a unifying view of machine translation as statistical and neural search in a large search space.  

The course covers the area of machine translation (MT) in its current breadth. Lectures will put a balanced emphasis on several important types of state-of-art-systems such as phrase-based machine translation and deep-syntactic machine translation, machine translation evaluation or methods.  We cover fundamental building blocks from linguistics, machine learning, algorithms, and formal language theory, showing how they apply to real and difficult problem in artificial intelligence. Finally, a detailed gist of emerging approaches in MT such as neural network is also included in the course.

• Statistical Machine Translation by Philipp Koehn, 2010
• Neural Machine Translation by Philipp Koehn, 2019

• Linguistic Fundamentals for Natural Language Processing by Emily Bender

• There will be 2 quizzes in total
• Similarly, there will be 4 homework in total

• User of mobile phones are strictly prohibited. In classroom or lab, if any student is caught using mobile phones, they may receive a warning first, and later may not be allowed to sit in class or lab.
• Attendance: Attending lectures and lab is mandatory. You will NOT be allowed to enter in class after 10 minutes. You are encouraged to be in class on time. If your attendance is less than 70%, there will be no compromise and you will be withdrawn from course an DZ grade will be assigned.
• Late submission NOT allowed: If you fail to submit your assignment/project on the specified time, you will NOT be able to get any marks.
• Plagiarism Unacceptable: You are not allowed to directly copy and paste any code available online for your project or homework, without understanding them. If you understood and used the online material, then you must provide reference to that website or resource from where you have copied it. Otherwise, it will be considered plagiarism and no marks will be given to you.