Probabilistic Graphical Models (EECS 474)

Fall 2016
Electrical Engineering and Computer Science Department
Northwestern University

Class Meets: 3:30PM-4:50PM TTh, Tech M177

Instructor: Doug Downey
Email: ddowney <at> eecs <dot> northwestern <dot> edu
Office Hours: Monday 4:00PM-5:00PM, Ford 3-345

Teaching Assistant (part-time): Nishant Subramani
Email: nishant.subramani23 <at> gmail <dot> com
Office Hours: Monday 4:00PM-5:00PM, Ford 3-345 (joint with instructor office hours)

Required Textbook: Koller and Friedman. Probabilistic Graphical Models: Principles and Techniques. The MIT Press.

Probabilistic graphical models are a powerful technique for handling uncertainty in machine learning. The course will cover how probability distributions can be represented in graphical models, how inference and learning are performed in the models, and how the models are utilized for machine learning in practice.

The course objective is for students to gain an understanding of how graphical models can be used to represent probability distributions, how to perform inference and learning in the models, and how the models are applied to natural language. Topics include directed and undirected graphical models, exact and approximate inference methods, and supervised and unsupervised parameter and structure learning.


The meetings for the course will consist of lectures regarding probabilistic graphical models and their application to machine learning tasks. Five homeworks will be assigned during this period (50% of the course grade) and two midterm exams will be given, one on 10-18-2016 and another on 11-17-2016. Each exam counts for 25% of the course grade. The remainder of the course will focus on lectures on statistical language modeling and a related final project (25% of the grade).


Homeworks, Lectures, etc.

Please see the course calendar.

Links to Bayes Net Packages