6.806/6.864 Advanced Natural Language Processing
The need to study human languages from
a computational perspective has never been greater. Much of the vast
amounts of information available today is in a textual form, requiring
us to develop automated tools to search, extract, translate, and
summarize the data. This course on natural language processing (NLP)
focuses exactly on such problems, covering syntactic, semantic and
discourse processing models, and their applications to information
extraction, machine translation, and text summarization. As a new
feature this year, the course will emphasize deep learning techniques
for NLP, introducing them in parallel and comparatively with more
traditional approaches to NLP.
Teaching staff (email@example.com):
(please read this page carefully before contacting the staff)
- Tuesdays and Thursdays, 1:00 pm--2:30 pm (32-123)
- Profs. Regina Barzilay, Tommi Jaakkola
Midterm (30% of grade):
- October 25, closed book, 1.5h, in-class
If you have a conflict, you must notify the staff by email
before October 6.
- You cannot pass the course without taking the midterm.
Four (4) homework assignments (20% of grade in aggregate):
- The assignments combine theoretical and empirical exercises and
require coding in python.
- They are always due 9am on the specified Tuesdays, unless otherwise specified.
- Late submission policy for homeworks:
Solutions will be
posted immediately after the deadline. As a matter of policy, late assignments are NOT
accepted, and correspond to getting zero on the assignment. The only exception to this rule is if you have a dean's
note. In this case, you must notify
the staff via email firstname.lastname@example.org. For students with a dean's note, the weight of their
missed assignment will be added to the midterm (for homeworks 1,2 and 3).
Missing homework 4 (after midterm) can only be
compensated by arranging an oral exam on the pertiment material.
- Collaboration policy for homeworks:
We encourage students to discuss assignments with other students and with the teaching staff to better understand the concepts. However, when you submit an assignment under your name, we assume that you are certifying that the details are entirely your own work and that you played at least a substantial role in the conception stage.
You should never use results (solutions, code) from other students (from
this year or previous years) in preparing your solutions unless you
developed the materials while working with other students in class.
Students should never share your solutions (or staff solutions) with other students.
Project (50% of the grade, see projects
for more details) :
- As the weight suggests, the project consitutes a significant learning component for this
course. The expected scope of the project differs between the graduate
(6.864) and undergraduate (6.806) versions of the course.
- As part of the project, you will design, implement, and evaluate a model for text processing
such as information extraction. The effort includes reading
additional relevant literature, implementing a baseline
model, analyzing its performance, proposing and developing an
improved version, and providing a comparative analysis.
- We will publish
a list of project templates but you can design your own (comparable)
version subject to the staff's approval.
The projects are carried out in groups of 4 students. We will
organize an event to facilitate matching.
- The projects require
regular meetings with the staff and involve clear milestones. You will submit a single joint write up (paper)
for the project but it must include a clear attribution of effort by the
members. Individual student contributions (pieces of the project) must be
clearly demarcated and identified.
- You will also prepare and present the project as a poster to the staff and
other members of the course. The poster presentations are held
during the last two lectures of the semester
Lecture scribing (opportunity to earn 5% grade bonus) :
- You can receive a 5% bonus (equivalent to one homework) by submitting (e.g., handwritten)
notes for half of the lectures during the semester. Instructions on
how to submit the notes will be provided separately.