Descriptif
Artificial intelligence, as a transversal discipline, plays a central role in our modern society, driving vital advances and amplifying efficiency, well-informed decision-making and general
practicality in our daily routines. This advanced master's course aims to provide students with a comprehensive understanding of the latest developments in Responsible AI. The course will explore various facets of Responsible AI, including interpretable AI, fairness in machine learning, robust machine learning, data privacy, and frugality. Students will delve into both theoretical foundations and practical implementations, equipping them with the skills to design and implement AI systems that are ethical, accountable, and aligned with
societal values.
Format des notes
Numérique sur 20Littérale/grade réduitPour les étudiants du diplôme M2 Data Science
Vos modalités d'acquisition :
Evaluation
Students will be asked to write a blogpost article on a paper recently published at a highranking conference (eg. ICLR, NeurIPS) related to the topic of the course.
A list of selected articles will be proposed by the lecturers. Blogposts will be published on the medium account of the master.
Example of blogposts:
https://iclr-blog-track.github.io/blog/
Programme détaillé
Organisation of the course (each session is 4 hours)
1. Interpretable AI - Florence d’Alché-Buc
2. Fairness in Machine Learning - Stephan Clemençon and Charlotte Laclau
3. Robust Machine Learning - Quentin Bouniot
4. Data Privacy - Yann Issartel
5. Frugal AI - Enzo Tartaglione and Florence d’Alché-Buc
6. Project supervision