Chloé-Agathe Azencott

This page presents the course Introduction to Machine Learning co-organised at Mines Paris PSL by Chloé-Agathe Azencott and Fabien Moutarde, with the help of Kelly (Hsiu-Wen) Chang Joly.

Starting in 2020, this course has been taking place every Fall in a semaine mixte 2A/3A i.e. as a week-long course open to second and third-year engineering students (equivalent to master's students).

Course presentation

In the age of data, various domains (Internet, marketing, logistics, biology, etc.) are accumulating enormous amounts of data, giving rise to the need for growing needs for automated tools that are able to exploit data of various nature. In particular, machine learning algorithms such as artificial neural networks, support vector machines or random forests can provide more powerful modeling and analyzes than classical linear statistical methods.

This course aims at providing an overview of these algorithms, as well as their theoretical and methodological framework, through a diversity of applications.

At the end of the course, students will be able to use state-of-the-art machine learning algorithms for

• supervised learning on tabular data;
• supervised learning on images or sequential data;
• clustering.

Each chapter will be covered by a lecture and a practical session. An afternoon of invited talks by researchers will illustrate current questions in machine learning in a diversity of application fields.

Prerequisites

Optimization and Data Science courses from the first year common core, or equivalent.

Evaluation

The evaluation is based on a written exam. Attending the invited talks is mandatory.

Schedule (2026)

Monday 9/14/2026

• morning: Introduction by Chloé Azencott
  • Empirical risk minimization, model selection and regularization
  • Mostly reminders from the first year common core Data Science course
  • TAs: Katia Antonenko and Mihaela Dimitrova
  • [slides (pdf)] [lab (zip)]
• afternoon: Multilayer perceptrons by Kelly Chang Joly
  • TAs: Mihaela Dimitrova and Wassil Lebsaira

Tuesday 9/15/2026

• morning: Deep learning and convolutional neural networks by Kelly Chang Joly
  • TAs: Mihaela Dimitrova and Wassil Lebsaira
• afternoon: Kernel methods and support vector machines by Chloé Azencott
  • TAs: Mihaela Dimitrova and Victor Gertner
  • [slides (pdf)] [lab (zip)]

Wednesday 9/16/2026

• morning: Decision tree and random forests by Fabien Moutarde
  • TAs: Stevan Le Stanc and Alexandre Olech
• afternoon: Recurrent neural networks by Kelly Chang Joly
  • TAs: Stevan Le Stanc and Alexandre Olech

Thursday 9/17/2026

• morning: Attention & transformers by Kelly Chang Joly
  • TAs: Alexandre Olech
• afternoon: Clustering by Chloé Azencott
  • TAs: Katia Antonenko and Victor Gertner
  • [slides (pdf)] [lab (zip)]

Friday 9/18/2026

• morning: Written exam
• afternoon: Symposium Current topics in machine learning
  • Thibault Faney (IFPEN) Scientific machine learning
  • Stevan Le Stanc Constrained ML for decision making under uncertainty
  • Mihaela Dimitrova Morphological neural networks
  • Alexandre Olech Monocular depth estimation with surface bias
  • Katia Antonenko Interpretable machine learning for epigenetics