Pattern Recognition

Course content

Upon completion of this module, students gain fundamental knowledge about methods and algorithms for classification of data. They are capable to select the appropriate means for real-world problems, to design a solution and to evaluate it.

•  Bayesian decision rule
• Quality metrics in pattern recognition
• Supervised learning with parametric distributions
• Supervised learning with non-parametric distributions, classification
• Linear discriminant functions, single-layer perceptron - Support vector machines (SVMs)
• Multi-layer perceptron, neural networks (NNs)
• Deep learning
• Unsupervised learning, clustering methods