EECS 595 001 FA 2021
Instructor: Joyce Chai (chaijy@umich.edu)
Time: Wednesday 1:30-3:00 pm; Friday 3:00-4:30 pm
Class Location: 1013 DOW
Course Description
The field of Natural Language Processing (NLP) is primarily concerned with computational models and computer algorithms to process human languages, for example, automatically interpret, generate, and learn natural language. In the past twenty years, the rise of the world wide web, mobile devices, and social media have created tremendous opportunities for exciting NLP applications. Recent advances in machine learning (e.g., deep learning) have also paved the way to tackle many NLP problems in the real world. This course provides an introduction to the state of the art in modern NLP technologies. In particular, the topics to be discussed include: syntax, semantics, discourse, deep learning for NLP, and their applications in information extraction, machine translation, and dialogue systems.
Text book
Speech and Language Processing, an introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition, third edition (draft) (Links to an external site.), by Daniel Jurafsky and James Martin, Prentice Hall (JM for short).
Optional: Neural Network Methods for Natural Language Processing, Yoav Goldberg, Synthesis Lectures on Human Language Technologies
Prerequisite
Proficiency in Python programming (programming assignments will be in Python). Knowledge and experience in machine learning is preferred.
Course Grades
The work in this course consists of four homework assignments and a final project. Each assignment may include a written portion and a programming portion.
- Homework assignments: 60%
- Final Project: 40% (Proposal: 5%; Related work: 10%; Presentation: 10%; Final Report: 15%)
Schedule of Topics and Assignments
Date |
Topics |
Reading and Assignments |
Sept 1 |
Introduction |
JM Chapter 2 |
Sept 2 |
Language Modeling with N-Grams |
JM Chapter 3 HW1 assigned |
Sept 8 |
Text Classification and Sentiment Analysis |
|
Sept 10 |
Logistic Regression and Neural Network |
JM Chapter 5 & 7 |
Sept 15 |
Vector Semantics |
JM Chapter 6 |
Sept 17 |
Neural Language Model |
JM Chapter 7.2 HW1 due, HW2 assigned |
Sept 22 |
POS and Recurrent Neural Networks |
JM Chapter 9 |
Sept 24 |
Contextual Embedding- Transformers |
|
Sept 29 |
Constituency Grammar and Parsing |
JM Chapter 12 |
Oct 1 |
Statistical Parsing |
JM Chapter 13, HW2 due, HW3 assigned |
Oct 6 |
Dependency Parsing |
JM Chapter 14 |
Oct 8 |
Meaning Representations |
JM Chapter 15 |
Oct 13 |
Semantic Parsing |
|
Oct 15 |
Semantic Roles and Selectional Restriction |
JM Chapter 19, 20, HW3 due, HW4 assigned |
Oct 20 |
Coreference Resolution |
JM Chapter 21, |
Oct 22 |
Discourse Coherence |
JM Chapter 22 |
Oct 27 |
Information Extraction |
JM Chapter 17 |
Oct 29 |
Question Answering |
JM Chapter 23. HW4 due |
Nov 3 |
Machine Translation |
JM Chapter 11, Final Project Proposal due |
Nov 5 |
Dialogue Systems and Chatbots |
JM Chapter 24 |
Nov 10 |
Recent Advances (1) |
|
Nov 12 |
Recent Advances (2) |
Final Project Related Work due |
Nov 17 |
Recent Advances (3) |
|
Nov 19 |
Recent Advances (4) |
|
Nov 24 |
Thanksgiving Break |
|
Nov 26 |
Thanksgiving Break |
|
Dec 1 |
Recent Advances (5) |
|
Dec 3 |
Recent Advances (6) |
|
Dec 8 |
Final Project Presentation |
|
Dec 10 |
Final Project Presentation |
|
Dec 16 |
Final Project Report Due |
All course materials, e.g., recorded videos, slides, assignments are in CANVAS
Course Policies
Homework
Homework must be turned in on the date that it is due, by 11:59 pm. The homework must be submitted electronically using Canvas and we will use the later timestamp to validate turn-in time. It is your responsibility to ensure that the homework has been uploaded successfully by the due date. Homework that is incorrectly uploaded will be subject to the associated late penalty. Late homework will be penalized 10% per day. Homework turned in after three days will not be accepted.
Also, note that any changes you make to the homework already submitted on Canvas count as a resubmission. If you make any changes to the assignment after the due date has passed you will be assigned a late penalty based on the number of days that have passed. For example, if you edit an assignment on March 5th and it was due on March 2nd, you will be assigned a 30% penalty (10% per day) as explained above.
Office Hours
The instructors will have regularly scheduled office hours each week. You are encouraged to make use of these to discuss aspects of the course including lecture material and homework problems. In cases where you cannot make office hours, contact the course staff to arrange an appointment.
Piazza
We have enabled Piazza to facilitate collaborative problem solving between students. It does not serve as constant on-demand access to course instructors. If you have pressing concerns, make sure to ask during lectures or office hours. Do not post homework solutions on Piazza.
Academic Honesty
Honor code
All homework submitted must be your own work. Review the Honor Code at the College of Engineering here: http://www.engin.umich.edu/college/academics/bulletin/rules (Links to an external site.)
The Honor Code is based on these tenets:
1. Engineers must possess personal integrity both as students and as professionals. They must be honorable people to ensure safety, health, fairness, and the proper use of available resources in their undertakings.
2. Students in the College of Engineering community are honorable and trustworthy persons.
3. The students, faculty members, and administrators of the College of Engineering trust each other to uphold the principles of the Honor Code. They are jointly responsible for precautions against violations of its policies.
4. It is dishonorable for students to receive credit for work that is not the result of their own efforts.
Among other things, the Honor Code forbids plagiarism. To plagiarize is to use another person's ideas, writings, etc. as one's own, without crediting the other person. Thus, you must credit information obtained from other sources, including websites, e-mail or other written communications, conversations, articles, books, etc.
Collaboration
We expect strict adherence to the Engineering Honor Code in all homework assignments. All problem sets (homework assignments) are to be completed on your own. You are encouraged to discuss ideas and techniques broadly with other class members, but all written works, whether in scrap or final form, are to be generated by you working alone unless otherwise expressly stated in the homework assignment. You are not allowed to sit together and work out the details of the problems with anyone. You are not allowed to discuss the problem set with previous class members, nor anyone else who has significant knowledge of the details of the problem set. Nor should you compare your written solutions, whether in scrap paper form or your final work product, to other students (and vice versa). You are also not allowed to possess, look at, use, or in any way derive advantage from the existence of solutions prepared in prior years, whether these solutions were former students' work products or copies of solutions that had been made available by instructors. Violation of this policy is grounds to initiate an action that would be filed with the Dean's office and would come before the College of Engineering's Honor Council. If you find any ambiguity about this policy, it is your responsibility to contact the course staff for clarification.
The final project can be a project with one person or a collaborative project with two people. More details on the final project will follow.
Special Accommodations
If you have disabilities or medical conditions that require some form of accommodations, please contact your instructor and the Office of Students with Disabilities at the start of the term so that arrangements can be made to accommodate you.
Notes: The instructor reserves the right to modify course policies and the course calendar according to the progress and needs of the class.