STAT 8305 - Pattern Recognition in Machine Learning

3 Credit Hours
Prerequisites: Math 6752 with a grade of C or higher.
Description
Pattern classification is about the design of algorithms to enable computers learn patterns from data. This course will provide a mathematical discussion and hands-on applications of Bayesian decision theory with known probability structure; Minimum error rate classification; Discriminant functions; Error bound estimation; Maximum-likelihood and Bayesian parameter estimation when full probability structure is not known; Discriminant analysis; Nonparametric classification; Information theory; Diffusion probabilistic model. The ability to program is helpful but not required.