Thematic School “Core-AI 2026”

March

• March 11 — Nirupam Gupta (University of Copenhagen)
  Robustness, Privacy, and Fairness in Distributed Learning
  (co-organized by ML-MTP) – LIRMM Seminar Room, Building 4 – 4:00 p.m.
• March 17 — Arnak Dalalyan (CREST – ENSAE)
  Generative models and GAN theory – LIRMM Seminar Room, Building 4 – 4:00 p.m.
• March 25 — Axel Ngonga Ngomo (University of Leipzig)
  Knowledge graphs – Jean-Jacques Moreau Lecture Hall, Building 2, St Priest – 3 p.m.

April

• April 1 — Marie-Jeanne Lesot (LIP6, Sorbonne University)
  Explainable AI: methods and risks – LIRMM Seminar Room, Building 4 – 4:00 p.m.
• April 9 — Hadrien Hendrikx (Inria Grenoble)
  Distributed optimization
  (co-organized by ML-MTP) – Jean-Jacques Moreau Lecture Hall, Building 2, St Priest – 4:00 p.m.
• April 15–16 — Mini-course (2 half-days)
  Dino Ienco & Giuseppe Guarino (INRAE Montpellier)
  Semi-supervised learning and domain adaptation – Jean-Jacques Moreau Lecture Hall, Building 2, St Priest – 4:00 p.m.

May

• May 6 — Céline Robardet (INSA Lyon)
  Understanding and Explaining Graph Neural Networks – Jean-Jacques Moreau Lecture Hall, Building 2, St Priest – 4:00 p.m.
• May 13 — Gérard Biau (Sorbonne University – Academy of Sciences)
  Learning with Physical Constraints (PINNs, Physics-informed Kernel Learning)
  (session may be extended) – Lecture hall at Triolet (room to be confirmed) – 4 p.m.

About Core-AI

Core-AI highlights the fundamental principles, theoretical methods, and methodological approaches that structure modern artificial intelligence.

The school offers a balanced science curriculum covering:

• Learning theory
• Robustness and confidence
• Distributed deep learning
• Generative models and kernels
• Large data structures (graphs, vision, remote sensing)
• Explainability and responsible AI

For more information, as well as to view other dates and events at the school, please refer to the registration form.

* By entering your email address, you agree to receive a weekly summary of the UM calendar by email and acknowledge that you have read ourprivacy policy. You can unsubscribe at any time using the unsubscribe linkor by contacting us by email.