Seminar in natural language processing. Topics vary by year. This is a seminar for advanced undergrads and graduate students who have prior background in natural language processing and would like to advance their knowledge and develop research skills.
With the advance of neural network approaches to NLP, several challenges have surfaced with their development and deployment. The 2021 iteration of the seminar will address two key challenges: a) interpretability, or how to open the black-box of neural NLP models, and understand their structure and their decisions; and b) robustness, or how to make models more robust to biases, spurious correlations, distribution shift, etc.
Interested students should fill out the following form: https://forms.gle/WBab9NS1mHDiaEDF6
Registration is now closed. If you want to join the wait list, fill out the form above.
Staff:
Yonatan Belinkov, instructor (belinkov@technion.ac.il)
Office hours: Monday 14:00-15:00. You can book a time here: https://belinkov236817.youcanbook.me/. Office hours will take place virtually via https://technion.zoom.us/j/7607219957 until further notice.
Learning outcomes:
By the end of this course, the student will be able to:- Present, discuss, and critically assess research papers in NLP.
- Define a novel or existing research question, design a solution, implement it, and write a research report.
Pre-requisites:
This is an advanced course that assumes background in NLP and machine learning, such as can be acquired via the following courses: 236299: Introduction to natural language processing, 236756: Introduction to machine learning, 236781: Deep learning on computation accelerators. Students without the official requirements will still be considered if they can demonstrate sufficient background.
Coursework:
- Read papers before each class.
- Post reading responses (1-2 paragraph long) about the weekly papers. The response should pose questions and comments about the paper that would be appropriate for discussion in class.
- Lead discussions of 1-3 papers throughout the course. The discussions would be led by groups of 2-3 students. Discussion leaders will read the responses from the entire class and synthesize them when leading the discussion.
- Complete a research project, in small groups of 1-3 students. The project includes three components:
- Literature review
- Project proposal
- Final paper
Language of instruction:
The course materials are in English. The language of instruction and discussion is Hebrew.
Grading:
The course grade will be comprised approximately as follows; the exact grade composition may change.
- Reading responses 15%
- Participation 10%
- Leading a discussion 25%
- Literature review 10%
- Project proposal 15%
- Final project paper 25%
Schedule:
The course meets weekly on Tuesdays at 10:30-12:30