One of the main challenges in building intelligent systems is the ability to learn and reason under uncertainty, and one of the most successful approaches for dealing with this challenge is based on the framework of Bayesian belief networks (and probabilistic graphical models more generally). Intelligent systems based on Bayesian networks are currently being used in a number of real-world applications including diagnosis, bioinformatics, channel coding, computer vision, text and social-network analysis, and data mining.
The objective of this class is to provide an in-depth exposition of knowledge representation, reasoning, and machine learning under uncertainty using the framework of Bayesian networks. Both theoretical underpinnings and practical considerations will be covered, with a special emphasis on constructing and learning graphical models, and on various exact and approximate inference algorithms. Additional topics include logical approaches to probabilistic inference, compilation techniques, sensitivity analysis, undirected graphical models, and statistical relational learning.
Grading will be based on Homework (30%), Midterm (30%) and Final (40%). Midterm and final will be closed book.
There will be 5 homeworks, which will all be announced on CCLE. You will have at least a week to complete them.
Adnan Darwiche. Modeling and Reasoning with Bayesian Networks. Cambridge University Press 2009.