The student has knowledge about basic statistics and mathematics, such as, e.g., maximum likelihood estimation, linear regression, binary classification, matrix algebra, optimization. The student can program in R or Python.

In this coursework we give a non-exhaustive overview of the basic principles of machine learning. In several classes we cover topics like, bias/variance trade off, simple linear regression and classification, cross validation and bootstrapping, unsupervised methods, feature selection methods, splines, random forests, and support vector machines. The theory is applied on a Kaggle competition. The course aims at junior level data scientists. Notion of programming and mathematics/statistics is mandatory.

-      A framework for machine learning
-      Simple supervised methods: Linear regression and Classification
-      Resampling methods, Model selection and Regularization
-      Moving beyond simple supervised methods:

      -      Regression splines
      -      Local regression
      -      Generalized additive models
      -      Tree-based method
      -      Support Vector Machines

- Unsupervised techniques