This 12h course develops important aspects of statistical modelling, which are particularly related to machine learning.
In practice, the courses and practicals will be structured in three blocks:
- Part 1: Basics of random modeling, maximum likelihood, mathematical formulation of the supervised learning problem
- Part 2: Simple linear regression (formulation, analytical definition of error limits, outliers detection). Multiple linear regression (formulation, bias-variance compromise, RIDGE and LASSO regularization, model selection, cross-validation)
- Part 3: Mixed models (formulation, factor significance test)