Paul G. Allen School of Computer Science & Engineering

CSE 599 D1 - Advanced Natural Language Processing - Spring 2020
Mon/Wed 1:30-2:50PM.
Zoom link is available in Canvas

Instructor: Luke Zettlemoyer, lsz@cs.washington.edu
Office Hours: by appointment

Course Objectives

This advanced course deeply explores advanced topics in NLP. The goal is to enhance students' knowledge about current techniques, challenges, and developments in different areas of NLP; to encourage discussions and collaborations among students; to improve students' analytical thinking and creativity; and to improve presentation and writing skills. In this class, students are required to read, think, present, and write about topics in modern NLP research.

Multi-lingual NLP

The vast majority of research in natural language processing (NLP) has focused exclusively on English language texts. However, there are thousands of languages in the world and recent advances in deep learning for NLP have introduced models that should, in theory, work for any language. In this class, we will review and discuss ideas in multi-lingual NLP, including but not limited to morphological analysis, character-level models, cross-lingual transfer, language model pre-training, and massively multi-lingual machine translation. We will read foundational and advanced papers on these topics, with a focus on more recent work. The goal is to acquire comprehensive understanding and insights into the emerging developments and challenges of these advanced topics, and to develop future research directions and original research ideas.

Topics and Schedule

The set of topics that we cover will be selected by the students during the first week of class. Some possible papers are listed here.

Class Activities and Evaluation

Students will be evaluated based on these four activities (% of final grade): (i) leading class discussions (40%), (ii) participation in the discussions (35%), (iii) 1 page paper proposal (10%), (iv) 4-5 page final paper, which can be a survey of any related area in NLP and can also include original research ideas or results for an addition extra credit of up to 15%.