The course will be run over eight weeks. The first six will include lectures, practicals and journal clubs in which students will be introduced to: week 1) the basic concepts in machine learning (e.g. feature extraction, feature selection, training testing and evaluation); week 2) classification (e.g. Naïve Bayes, Logistic Regression and Support Vector Machines); week 3) multivariate regression (e.g. Ridge Regression) and modelling of neural response data; week 4) bio-inspired machine learning approaches including (deep) neuronal networks and evolutionary computation; week 5) advanced concepts in feature selection (e.g. bagging); week 6) unsupervised learning. In the last two weeks the students will be given a data set on which they will have to apply the concepts they have previously acquired by implementing a classification/regression/clustering analysis pipeline.
To introduce students to the basic and advanced concepts of machine learning and multivariate statistics. The course will introduce supervised (classification and regression) and unsupervised (clustering) learning
Pattern Classification – R.O. Duda, P.E. Hart, D.G. Stork; John Wiley & Sons (2012) Various articles for the Journal clubs will be selected by the tutors