Description
This is an introduction to probabilistic models and inference algorithms, which constitute a common foundation for many methodologies in machine learning and related fields (e.g. computer vision, natural language processing, and data mining).
The course begins with a detailed exposition of probabilistic graphical models and exponential families, and then proceeds with various inference and estimation methods for probabilistic models, including both optimization-based methods (e.g. belief propagation and variational Bayes) and sampling-based methods (e.g. Markov Chain Monte Carlo). Finally, it will cover nonparametric models, including those based on Gaussian processes and Dirichlet Processes. If time permits, it will touch more advanced topics, e.g. the combination of graphical model and deep neural networks.
The course begins with a detailed exposition of probabilistic graphical models and exponential families, and then proceeds with various inference and estimation methods for probabilistic models, including both optimization-based methods (e.g. belief propagation and variational Bayes) and sampling-based methods (e.g. Markov Chain Monte Carlo). Finally, it will cover nonparametric models, including those based on Gaussian processes and Dirichlet Processes. If time permits, it will touch more advanced topics, e.g. the combination of graphical model and deep neural networks.
General Information
Time
Mon 1:30 pm - 2:30 pm
Wed 3:00 pm - 4:30 pm
Wed 3:00 pm - 4:30 pm
Venue
SHB 833
Name | Office Hours | |
---|---|---|
Dahua Lin | When? Where? | |
DaiBo | When? Where? |