MTH 6312 is a graduate course targeting Statistical Learning and Data Mining from methodological point of view. The official page of the course on the Polytechniques graduate course web list is here. If you had problems with taking the course, contact our graduate office at Ecole Polytechnique or send an email to vahid d0t partovinia at-sign polymtl d0t ca. The course will be given every fall. Lectures will appear at one of the Ecole Polytechnique de Montreal's buldings. The teaching language is flexible (English or French) depending on participants.
We discuss the new supervised, semi-supervised and unsupervised methods.
Main subjects include:
Loss Function and Optimization
Bias-variance trade-off, model selection criteria
Bayesian linear models
Lasso regression, Least angle regression
Nonparametric regression and smoothing splines
Discrimination techniques (logistic regression, linear and quadratic discriminants, nonparametric logistic regression, mixture of discriminants)
Separating hyperplanes and support vector machines
Additive trees, MARS, and CART
EM algorithm and applications in semi-supervised learning
Bagging and boosting
Bayesian parametric clustering
Bayesian functional clustering
Bayesian nonparametric models and Dirichlet processes
Do you look for more fun with machine learning? My research team figures out which one is more likely to talk like you
Mitterand or Chirac?