_MPII Home Page_

AG 3: Teaching

Up to: Research Units Building 46.1 Algorithms and Complexity Group
Programming Logics Group
Bioinformatics Group
Computer Graphics Group
People
Projects
Offers
Teaching - Selected
Talks and Events
Publications
Software
Useful Links

The Elements of Statistical Learning

News:

Lecturer: Thomas Lengauer

Tutor: Jörg Rahnenführer

Course language: The course is taught in English

Time and location:

Course: Thursday 13-16, Building 46, Room 024; starting on Thursday, October 24, 2002.
Tutorial: Wednesday 13.30-15, Building 45, Room 014; starting on Wednesday, October 30, 2002.
Calling hours:
Thomas Lengauer: TBA, Building 46, Room 502 (...and after each lecture).
Jörg Rahnenführer: TBA, Building 46, Room 520.

Target Group:

Advanced students in math, computer science, science students with substantial mathematical background
Prerequisites: Vordiplom in Math. or Computer Science or equivalent. Basic knowledge in statistics.

Literature:

Hastie, Tibshirani, Friedman: The Elements of Statistical Learning, Springer 2001. The readers of the course are encouraged to acquire this book.

Content:

Statistical Learning is a fundamental methodology that is central to data mining. As such the method has many fields of application, including computational biology, vision and scene analysis, geographic databases, robotics etc. The aim of the course is to present an overview of statistical learning methods that enable the listeners to go into depth on specific topics of their choice. This means that we will not present all details of the chapters. Nevertheless, all chapters will be touched, in order to span the whole field within one semester, at least at some level. The course will be patterned after the introductory text mentioned above. Each lecture will cover one chapter of the book. Thus, the topics of the lectures will be:

  1. Introduction
  2. Overview of Supervised Learning
  3. Linear Methods for Regression
  4. Linear Methods for Classification
  5. Basis Expansion and Regularization
  6. Kernel Methods
  7. Model Assessment and Selection
  8. Model Inference and Averaging
  9. Additive Models, Trees, and Related Methods
  10. Boosting and Additive Trees
  11. Neural Networks
  12. Support Vector Machines and Flexible Discriminants
  13. Prototype Methods and Nearest Neighbors
  14. Unsupervised Learning
The tutorials will deepen the discussion and treat homework problems. Homeworks will cover both theoretical and programming exercises. We will use the statistical programming package R for the programming exercises.

Requirements for the course certificate (8 credit points):

50% of the homework, final exam (probably oral)

Course Material:

restricted access


Home / Research Units / AG 3: Home Page / Teaching | Back to the top of this page
[an error occurred while processing this directive] [an error occurred while processing this directive]
Document last changed on Wednesday, 04 September 02 - 09:55