Course Information

: 2+1
Lecturer: Thomas Breuel
  • Hastie, Tibshirani, Friedman: Elements of Statistical Learning [ESLI] (download)
  • Duda, Hart, Stork: Pattern Classification (library)
Course Language: English (lectures), English or German (exams, exercises)
Lectures: Mo, 08:15 - 09:45  48-462 
Exercises: Wed, 16:00 - 17:00 48-211 (attend even if there is no homework)
Start:  26.10.2009
Written Examination: March 1 2010, 10:00 - 11:30, 11-262, (second date: July 28, 2010, nobody participated on that day). Next exam November 8, 2010 Please register with "Prüfungsamt"

All the Sage notebooks are now accessible as PDF files from this site.

For preparing for the exam, you may also find the lecture notes from previous years useful.

Lecture Topics

  1. nearest neighbor classification and nearest neighbor algorithms
  2. feature extraction and common feature types
  3. neural networks, gradient descent
  4. RBF networks and interpolation
  5. perceptrons and support vector machines
  6. k-means clustering, Gaussian mixtures, and semi-supervised learning
  7. VQ, principal components analysis and compression
  8. hierarchical clustering, dimensionality reduction
  9. decision trees
  10. pattern recognition with graphs
  11. generative data models and model-based classification
  12. Bayesian decision theory
  13. ML and Bayesian parameter estimation
These algorithms are widely used in many areas of computer science and will be illustrated on a variety of problems, including text and data mining, network security, image classification, OCR, and handwriting recognition. 

The emphasis of the course is on a working understanding of algorithms and their properties, but also covers some important theoretical results.

The course is accompanied by practical exercises in Numerical Python.