UE: Machine learning for scientific data – Apprentissage pour les données scientifique
| Responsable(s), établissement(s) | Balázs Kégl, LRI/UPSud Alexandre Gramfort, Telecom ParisTech |
| Adresse(s) mail | kegl@lri.fr alexandre.gramfort@telecom-paristech.fr |
| Lieu principal d’enseignement | UPSud |
| ECTS | 2.5 |
| Nombre d’heures total | 21 |
| Cours | 12 |
| TD | 0 |
| TP | 9 |
| Objectifs | Students will put their basic machine learning and data analysis knowledge to test for solving practical data science problems in scientific or industrial applications. Typically, we will treat three/four concrete problems coming from scientific or industrial applications (e.g., brain imaging, astrophysics, biology/chemistry, ad placement, insurance pricing). We describe and formalize the motivating problem, discuss the possible solutions, choose one, and assist the students in solving the problem. We will provide data and advise students on their choice of tools. A the end of each 3 CM + 3 TP session, students will be evaluated based on their solutions. |
| Prérequis | basic algorithmics, programming, and at least one machine learnig, statistics, signal processing, or data analysis course |
| Language | English (if possible) |