DOCUMENTS
The first document is about the basics of Probability Theory which is a prerequisite for the Master: PDF PROBA
Understanding of the basics of Probability Theory represents an important and necessary step in view of completing the program. Indeed, most of the courses, for instance, Machine Learning or Models for Dependent Data, will heavily rely on these notions. Exercises are provided so that the students may check their understanding. Of course, some personal research and complementary readings are still recommended for complete understanding.
The second document (more brief) is about Linear Algebra. For most of the students entering the program, it represents a reminder but still, it is important to refresh these notions because of its usefulness for instance in principal component analysis (PCA) and linear models. In particular, the notions of linear vector space, matrix inversion, eigenvalues, and basis decomposition in Hilbert spaces are crucial. PDF LINALG
After reading the two previous documents, one might be interested in Linear Regression which illustrates perfectly the notions of Probability (first documents) as well as Linear Algebra (second document). PDF OLS
Here is a small TEST to check your training.
At the beginning of the first semester, some class hours will be dedicated to checking the students’ understanding, answering questions, and clarifying some points.
OTHER RESOURCES
Several good books are accessible online for free (links below were valid in March 2022).
Reading these is not a requirement but they provide interesting complementary information. The first and second books are about Math and might be used as a complement to the previous documents.
https://mml-book.github.io/book/mml-book.pdf
https://probml.github.io/pml-book/book1.html
The third book is more advanced as it deals with ML. A full course on this topic will be taught during the program.
https://hastie.su.domains/ISLR2/ISLRv2_website.pdf
TRAIN YOUR CODING SKILLS
There is plenty of relevant formation about Python and R available online. Each student should be familiar with both before the beginning of the program. We recommend using COURSERA; DATACAMP and Fun-MOOC (for which many courses are available in French).
For instance:
https://www.fun-mooc.fr/en/courses/machine-learning-python-scikit-learn/
https://www.coursera.org/learn/probability-intro
https://app.datacamp.com/learn/courses/intro-to-python-for-data-science
https://app.datacamp.com/learn/courses/free-introduction-to-r
NB: Most online courses provide free access to the first chapter which would be enough for practicing before the program starts.
