Pattern Classification and Machine Learning

Om kursen

Course content:

Introduction: outline, "road map" for classification methods

  • Signal characterization, discriminative features, feature normalization, examples.
Supervised learning methods
  • ML, Bayesian parameter estimation for pattern classification;
  • Statistical learning and boosting methods (Support Vector Machines and Adaboost);
  • Bayesian and belief networks;
  • GMMs and EM algorithm.
Unsupervised learning methods
  • ML parameter estimation for component densities from the mixture density;
  • Clustering methods (K-means, mean-shift);
  • Topic models and BoW (bag of words);
Automated learning of features
  • NNs and deep learning.

Mer information

Irene Gu E-mail:


Course book/material:

  1. Richard O. Duda, Peter E. Hart, David G. Stork, Pattern classification, John Wiley & Sons, 2nd Ed.
  2. Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer.
  3. David Barber, Bayesian reasoning and machine learning, Cambridge university press, 2012.
  4. Related state-of-the-art publications and tutorial materials
  5. Matlab toolbox for PR.

Home exercises and Exam:
  • Solving several theoretical problems
  • Matlab programs on several selected topics and presentations
  • Written exam: assignment of problem solving and programming tasks.


Irene Gu