Machine Learning

Supervised machine learning methods for regression and classification. Overview of basic notions in machine learning; linear regression and its extensions; logistic regression, generative models for classifications and generalized linear models; cross-validation and the bootstrap; model selection and regularization, including subset selection, ridge and lasso, and dimension reduction methods; polynomial and spline regression, generalized additive models; tree methods, including bagging, random forests, boosting and Bayesian decision trees; support vector machines; deep learning architectures, including convolutional and recurrent neural networks, as well as practical implementation aspects. Extra topics include unsupervised learning, conformal prediction and multiple testing. The course has a computing component in Python.