Statistics (Statistical Learning Theory) OpenCourseWare: A Free Online Grad-Level Statistics Course by MIT

Published Jan 07, 2009

RSS Feed

The Massachusetts Institute of Technology (MIT) presents 'Topics in Statistics: Statistical Learning Theory,' a free OpenCourseWare. This graduate-level course is appropriate for students interested in mathematics and statistics, who are interested in furthering their understanding of the generalization ability of several popular machine-learning algorithms. Recommended prerequisites include courses in probability, statistical learning theory and applications or machine learning.

Topics in Statistics: Statistical Learning Theory: Course Specifics

Degree Level Free Audio Video Downloads
Graduate Yes No No Yes

Lectures/Notes Study Materials Tests/Quizzes
Yes Yes No

Topics in Statistics: Statistical Learning Theory: Course Description

The 'Topics in Statistics: Statistical Learning Theory' course originally presented by Professor Dmitry Panchenko of MIT explores popular machine-learning algorithms. These including support vector machines, boosting and neural networks. Specific topics of study include the Vapnik-Chervonenkis theory, concentration inequalities in product spaces and one-dimensional concentration inequalities. These topics are presented through a series of 42 lectures accompanied by recommended readings. Two problem sets challenge students in their understanding of these topics. Students are encouraged to complete courses in probability, statistical learning theory and applications or machine learning before embarking on this graduate-level course.

The OpenCourseWare includes lectures, problem sets and a list of recommended readings. To download materials needed complete the course, visit the statistics in statistical learning theory course web page.

Featured School Choices: