- Department of Cybernetics
I assume the student has basic mathematical background. Being familiar with probability theory and statistics is a plus.
- Syllabus of lectures:
Formulation of the basic tasks solved in pattern recognition. Bayesian and non-Bayesian tasks. Two special useful statistical models. Conditional independence of features. Gaussian models. Strightening of the feature space. Estimation of probabilistic models. Parametric and nonparametric methods. Experimental evaluation of classifiers. Receiver operator curve (ROC). Learning in pattern recognition. VC dimension. Estimate of the needed length of the training sequence. Learning in pattern recognition. VC dimension. Estimate of the needed length of the training sequence. Linear classifier. SVM classifier. Kernel methods. Unsupervised learning. Cluster analysis. EM (Expectation Maximization) algorithm. Intro to structural methods embedded into the statistical framework. Recognition of Markovian sequences. Structural pattern recognition, a classical approach. Experiences learned from practial implementations of pattern recognition methods.
- Syllabus of tutorials:
The subject does not have labs or exercises. Students write a training paper with the help of the lecturer.
- Study Objective:
- Study materials:
Schlesinger M.I., Hlavac V.: Ten lectures from statistical and structural
pattern recognition, Kluwer Academic Publishers, 2002.
Duda R.O., Hart P.E., Stork D.G.: Pattern Classification, John Wiley and
- Further information:
- No time-table has been prepared for this course
- The course is a part of the following study plans: