logo Insalogo Insa

Machine Learning

Objectives


At the end of this module, the student will have understood and be able to explain (main concepts):
-          Principle of penalised likelihood for balancing variance vs. bias of a model
-          Optimal variable selection and/or regularisation
-          Selection and validation of statistical models
-          Unbiased predictive error estimation.
 
The student will be able to:
-          Process very high dimensional data from coming various domains (biology, medicine, insurance, marketing, finance...) with classical statistical software (SAS, R).
-          Estimate statistical models for regression or classification purposes with different approaches (general linear model, PLS, k-nearest neighbours, neural network, binary tree, support vector machine, bagging, boosting...)
-          Estimate various kinds of predictive errors,
-          Optimize model selection in order to build and validate a best predictive model.
-          Apply this strategy to a real data set

Needed prerequisite


Elements of statistical modelling
Statistical exploration and software'.s

Form of assessment

The evaluation of outcome prior learning is made as a continuous training during the semester. According ot the teaching, the assessment will be different: as a written exam, an oral exam, a record, a written report, peers review...