The course consists of lectures and programming practicals. Students will need to present the outcome of the practical work and take a written exam in the end.
The following is copied from LSF:
Content: The lecture builds on WP4 and WP5 and aims to provide an overview of important advanced data-analysis methods derived from neuroscientific principles and/or applied in neuroscience. The following topics will be covered: Perceptron and linear separability, logistic regression, cross validation, multilayer networks, backpropagation, empirical risk minimization, regression, density estimation, regularization, support-vector machines, optimization with constraints, kernel trick, bootstrapping, clustering, Bayesian networks, hidden Markov models, dimensionality-reduction techniques, time-series-analysis methods (multitaper estimators, wavelet decomposition, multivariate spectral estimators, hybrid time series), circular statistics.
Learning outcomes: The students will be able to understand the functioning and applicability of important advanced data-analysis methods derived from neuroscientific principles and/or applied in neuroscience, and be able to interpret results derived with these methods.
- Teacher: Christian Leibold
- Teacher: Anton Sirota