Probabilistic Machine Learning. 3 credits

Catalog description

Machine learning relies on probabilistic models to train computers to perform intelligently certain tasks such as natural language processing, image recognition, robot navigation, and resource allocation in communication networks. This course covers powerful probabilistic methodologies that have proven useful in machine learning. Both supervised learning and unsupervised learning will be discussed. We begin with basic statistical inference based on Bayesian decision theory and then discuss probabilistic models and inference methods for machine learning. These include hidden Markov models, graphical models, Bayesian networks, regression, kernels, Markov Chain Monte Carlo, and particle filters. Offered by Electrical & Comp. Engineering . May not be repeated for credit.