Skip to content


Course Description

Statistical methods and machine learning have had significant impact in science and industry. Recently, nonparametric methods have transformed these fields by significantly increasing the modeling capabilities for regression, density estimation, and classification. This course will explore a number of flexible nonparametric models including splines, kernel methods, regression trees, random forests, etc. We will also touch upon some modern ideas from Bayesian nonparametrics (e.g., Gaussian processes, Dirichlet processes); Bayesian interpretations of other models will be presented when possible. The course addresses both theoretical and practical aspects of these methods.

Throughout the quarter, students will gain experience with implementing the presented methods through a series of homework assignments. In addition, each student will complete a course project that examines a method beyond what was presented in class.


  • Homework (60%)
  • Project (40%)