Intended learning outcomes *
After the course, the students should be able to
*explain, derive and implement a number of models of supervised and unsupervised learning,
*analytically demonstrate how different models and algorithms relate to one another,
*explain strengths and weaknesses for different models and algorithms,
*choose appropriate model or strategy for a new machine learning task.
More specifically, regarding methodologies the student should be able to
*explain the EM-algorithm and identify problems where it is applicable,
*explain the terminology for Bayesian networks and topic models and apply these on realistic amounts of data,
*explain and derive boosting algorithms and design new boosting algorithms with different cost functions,
*explain and implement methods for learning of feature representations from various types of data.