CSE P 517 - Natural Language Processing

CSE P 517 -- Natural Language Processing [Winter 2021]

Lectures: MW 6:30pm-7:50pm (zoom info available at Canvas)

Teaching Crew

Personnel	Contact	Office Hours
Instructor: Yejin Choi	yejin at cs dot washington dot edu	Mon & Wed 8pm - 8:30pm
TA: Xiujun Li	xiujun at cs dot washington dot edu	TBD (zoom info available at Canvas)
TA: Jae Sung (James) Park	jspark96 at cs dot washington dot edu	TBD
TA: Peter West	pawest at cs dot washington dot edu	TBD
TA: Ximing Lu	lux32 atcs dot washington dot edu	TBD

Week	Dates	Topics & Lecture Slides	Notes (Required)	Textbook & Recommended Reading
1	Jan 4, 6	Introduction [slides] Probabilistic Models of Words: N-gram Language Models (LMs) [slides]	LM	JM-ed3 Ch 3 JM-ed2 Ch 4.1-4.7
2	Jan 11, 13	Probabilitic Models of Sequences: Hidden Markov Models (HMMs) [slides]	HMM	JM-ed3 Ch 8 JM-ed2 Ch 6.1-6.5
3	Jan 20 (Jan 18: MLK day)	Probabilitic Models of Sequences: Hidden Markov Models (HMMs) & EM [slides]	HMM, EM, Forward-backward	JM-ed3 Ch 8 JM-ed2 Ch 6.1-6.5
4	Jan 25, 27	Learning (Feature-Rich Models): Log-Linear Models, Maximum Entropy Models [slides]	LogLinear, MEMMs	JM-ed3 Ch 5, 8 JM-ed2 Ch 6.6-6.8
5	Feb 1, 3	Learning (Structural Graphical Models): Conditional Random Fields (CRFs) [slides] Guest lectures by Max Forbes on Verb Physics and Factor Graphs and Social Chemistry 101	MEMMs, CRFs	JM-ed3 Ch 8 JM-ed2 Ch 6.6-6.8
6	Feb 8, 10	Neural Language Models: Basics [slides]		JM-ed3 Ch 7, 9
7	Feb 17 (Feb 15: Presidents Day)	Neural Language Models: Transformers [slides]		JM-ed3 Ch 7, 9
8	Feb 22, 24	Neural Language Generation and De-generation [slides]		JM-ed3 Ch 24
9	Mar 1, 3	Neural Knowledge and Reasoning: On Commonsense Intelligence [slides]
10	Mar 8, 10	New Challenges with Neural Models: Interpretability, Bias, and Ethics

Recommended: [JM-ed3] Dan Jurafsky and James H. Martin, Speech and Language Processing: "An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition", draft chapters of the third edition available online
Optional: [JM-ed2] Dan Jurafsky and James H. Martin, Speech and Language Processing: "An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition", Prentice Hall, Second Edition, 2009.
Optional: [GBC] Ian GoodFellow, Yoshua Bengio, and Aaron Courville, "Deep Learning" (free online book available at deeplearningbook.org )

Available at Canvas

The grade will consist of

5 programming assignments (75%)
one self-defined final project based on either a replication of a paper, or a literature survey, or an open-ended empirical study (15%)
course participation (in-class and online forums) (10%)

Assignments must be done individually unless otherwise specified. Final projects can be done as a team of 1-3 people. You may discuss the subject matter with other students in the class, but all final answers must be your own work. You are expected to maintain the utmost level of academic integrity in the course.
Each assignment may be handed in up to three days late, at a penalty of 20% of the maximum grade per day. You have 5 penelty-free late day credits that you can use at any time during the quarter. Above 20% substraction will apply only after you have used all your late day credits. Being late by a partial day (e.g., 1 hour) will be rounded up to 1 full day.