Michèle Sebag
Also known as Michele Sebag
Michèle Sebag is a machine-learning researcher whose work includes causal modeling, preference learning and optimization. Sebag studied mathematics at École Normale Supérieure, then worked at Thalès, consulted and pursued doctoral research connecting learning with numerical engineering. She joined CNRS in 1991. She led the inference and machine-learning group at LRI from 2001 and co-founded the TAO project with Marc Schoenauer in 2003. Her projects include workplace conditions, company performance, scientific networks and recommendations for job seekers. These projects connect learning methods with questions about decisions and social systems.
- Now at
- CNRS and Université Paris-Saclay
- Works
- 525
- Citations
- 6,003
- h-index
- 43
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
Before 1991
Mathematics and engineering
Sebag studied mathematics at École Normale Supérieure and worked in industry at Thalès. She consulted and pursued doctoral research combining machine learning, data analysis and numerical engineering.
-
1991–2003
Building research groups
She entered CNRS as a research fellow in 1991. In 2001 she took charge of the inference and machine-learning group at LRI. In 2003 she founded the TAO project with Marc Schoenauer.
-
2015–2019
Research community service
She headed the steering committee for the European machine-learning and knowledge-discovery conference from 2015 to 2019. She also gave talks on recommendations, representation learning and causal modeling in this period.
-
As of October 2026
Learning about social systems
Her interests include causal modeling, preference learning and surrogate optimization. Her projects examine workplace quality and company performance, scientific social networks, and recommendations for job seekers.
Sources and credits
Photo: Editors: Luc De Raedt, Siegfried Nijssen, Barry O’Sullivan, and Michele Sebag, via Wikimedia Commons. License: CC BY 3.0. Source: commons.wikimedia.org. Cropped to a square and resized. Ask us to remove or replace this photo.
