نظرة عامة على البرنامج

Program Overview

The University Program in Statistics and Machine Learning is a 6-credit unit of instruction that provides students with a comprehensive understanding of statistical theory and machine learning techniques.

Program Content

The program is divided into two main parts: Statistics and Machine Learning.

• I. Statistics
  • Partie 1: Introduction to minimax theory
  • Partie 2: Parametric models, maximum likelihood method, and Bayesian estimation
  • Partie 3: Non-parametric models, density estimation, non-parametric regression, kernel methods, projection methods, and piecewise polynomial methods
• II. Machine Learning
  • Partie 1: Supervised classification, linear methods, regularization, resampling methods, and random forests
  • Partie 2: Unsupervised classification, K-means, K-medoids, and mixture models
  • Partie 3: Topological data analysis in machine learning and the machine learning pipeline

Skills to be Acquired

The program aims to equip students with the following skills:

• Learn the tools and methods of modern mathematical statistics
• Understand the mathematical and algorithmic foundations of some machine learning methods

Language of Instruction

The language of instruction for this program is French.

Bibliography

The program's bibliography includes:

• Introduction to non-parametric statistics by A.B. Tsybakov
• Fundamentals of statistical learning by Sylvain Arlot
• Statistical learning and big data by Myriam Maumy-Bertrand, Gilbert Saporta, and Christine Thomas-Agnan

Prerequisites

The prerequisites for this program are:

• Integration
• Statistics
• Basic programming in Python

Volume of Instruction

The program consists of:

• 36 hours of lectures
• A total of 36 hours of instruction

Pedagogical Responsibilities

The pedagogical responsibilities for this program are held by:

• FREYERMUTH Jean-Marc
• LEPSKI Oleg

APOGEE Codes

The APOGEE code for this program is SMACUH6T [ELP].

Related Formations

This program is used in various formations, which can be searched for further information.