Statistical Learning: Modeling, Prediction and Computing
STAT 538 Winter Quarter 2012

 Home Course Description Books and other resources Class mailing list   UW Statistics Announcements Old exams with solutions are posted here: exam 11, sol exam 11,exam 10,sol exam 10 The exam will be open-book. You can bring any materials on paper but NO ELECTRONICS. Office hour today Tuesday in PDL B-321 4:30-6pm. NO office hour on Wednesday. Who is this class for? This class is the second of a sequence intended for statistics and biostatistics students (the previous course being STAT 535) with an interest in statistical learning, algorithms for statistical inference, and models for multidimensional data, as well as for other graduate students with an interest in statistics, algorithms and computing. The focus of the present course will be on supervised learning, and its connections to optimization. A more detailed description of the topics is given here. The grade is based on homework (60%), final exam (25%) and class participation (15%) (approximately). Prerequisites EITHER STAT 535 OR A course in probability, including basic notions of multivariate analysis (conditional probability, marginals, expectation, variance) Notions of statistics: Maximum Likelihood Estimation, MAP estimation, priors, likelihood, estimating parameters of usual distribution (normal, multinomial) Calculus: partial derivatives, the chain rule, vectors and matrices, matrix multiplications, gradient Algorithms and data structure at a basic level (arrays, lists, sets, O( ) notation). Medium ability with a computer programming language (like C, C++, Java or Matlab, Splus, R) at the level of STAT 534 Instructor: Marina Meila   mmp@stat.washington.edu Lectures: Tuesdays & Thursdays 11:30 - 12:50 in LOW 101 Office hours: Wednesdays 2-3 in Padelford B - 321 (tentative) Course home page: http://www.stat.washington.edu/course/stat538/winter12 (this page) Contact the instructor at: mmp@stat.washington.edu