Teaching: Courses & Tutorials
In Fall 2024, I am teaching UCI CS177: Applications of Probability in Computer Science.
Applied Bayesian Nonparametrics
Bayesian nonparametric (BNP) models define distributions on infinite-dimensional spaces of functions, partitions, or other combinatorial structures. They lead to flexible, data-driven unsupervised learning algorithms, and models whose internal structure continually grows and adapts to new observations.
- Tutorial at CVPR 2012: Hierarchical BNP models, learning & inference algorithms, temporal & spatial dependencies, applications to image & video analysis.
- Brown CS295P (Fall 2011): Graduate course on BNP models, inference, & applications.
- Erik Sudderth's PhD thesis (Chap. 2) has a tutorial on hierarchical Dirichlet processes.
- Peter Orbanz has collected a list of other tutorials on Bayesian nonparametrics.
Probabilistic Graphical Models
Graphical models enable scalable probabilistic modeling by decomposing complex distributions into local interactions. These graduate courses explore state-of-the-art variational and Monte Carlo methods for statistical learning with probabilistic graphical models.
- UCI CS274B: Learning in Graphical Models covers conceptual foundations needed for research advances, and was last taught in Spring 2024.
- UCI CS275P: Machine Learning with Generative Models takes a more application-oriented view of graphical models (and generative models more broadly), and was last taught in Spring 2024.
- Brown CS242: Probabilistic Graphical Models was taught from 2013 to 2016.
- Brown CS295P (Spring 2010) was an earlier seminar-style course on graphical models.
- Erik Sudderth's PhD thesis (Chap. 2) reviews graphical models & exponential families.
- Erik Sudderth & Bill Freeman wrote a tutorial on signal & image processing with belief propagation.
- Michael Jordan wrote an introduction to graphical models.
Machine Learning
How can artificial systems learn from examples, and discover information buried in massive datasets? This advanced undergraduate course explores the theory and practice of statistical machine learning, focusing on computational methods for supervised and unsupervised data analysis.
- UCI CS178: Machine Learning & Data Mining was last taught in Winter 2021.
- Brown CS142: Machine Learning was taught from 2009 to 2015.
- Machine learning textbooks: Bishop, Murphy, Barber, Hastie & Tibshirani & Friedman
Probability in Computer Science
Probabilistic methods and statistical reasoning play major roles in machine learning, security, web search, robotics, program verification, and more. This introductory course on probability and statistics emphasizes computational methods and computer science applications.
- UCI CS177: Applications of Probability in Computer Science was last taught in Fall 2024.
- Brown CS145: Probability & Computing was taught in 2015 and 2016.
- Probability textbooks: Harchol-Balter, Bertsekas & Tsitsiklis, Pitman, MacKay
