21/1 – Statistical Machine Learning

Main themes and topics: The course is intended to introduce statistical learning techniques and their implementation in the open source programming environment R. It gives an overview of important concepts, for instance regression and classification problems, the bias-variance trade-off and supervised vs unsupervised learning. The course covers methodologies for regression, cross-validation, variable selection and clustering. The course can be useful in almost any field of study: analysing and drawing meaningful conclusions from data is an important skill in many disciplines. The course can be particularly interesting for students with an interest in statistics and research methodology, but is not intended for those who have a passion for theories and proofs. A good mathematical aptitude and comfort with algebraic manipulations is assumed.