Classical Machine Learning

The aim of the course to introduce student to the basics of classical Machine Learning algorithms and concepts.

1. Linear regression, linear basis function models: modell formulations, training, feature selection, statistical tests

2. Basics of machine learning: train-, validation and test sets, bias-variance trade-off, regularization

3. Linear models for classification: Fisher discriminant, perceptron, generative models,

4. Evaluation of machine learning model performance, ML pipelines

5. Decision trees, random forests: modells, basics of information theory, pruning methods, building random forests

1. Summary and presentation about an optimization algorithm (10 points) – week 4.

2. Midterm oral exam (10 points) – week 7.

2. Summary and presentation about a special supervised algorithm (30 points) – week 13.