My research interests lie at the intersection of artificial intelligence and decision theory. My high-level goals are to help build decision-support systems (recommendation systems, decision aid, search...) and automatic decision systems (autonomous agents, robots, traders...). I am particularly interested in decision making under uncertainty (study of non expected utility models) and in exploiting those sophisticated models in optimization and sequential decision making problems (study of new algorithms). As a Ph.D student, I did some works (mainly axiomatizations) in qualitative decision making and unified many decision models in an algebraic framework. Besides, I proposed (joint work with Patrice Perny and Olivier Spanjaard) the general model of algebraic Markov decision processes and gave some sufficient conditions for the application of the backward induction. My research work now mainly focuses on sequential decision-making in contexts where not all preferential information is available (how a human can "teach" an agent to do a certain task?) and/or when the environment may evolve.
Keywords: Artificial intelligence; Algorithmic Decision theory; Markov decision processes/reinforcement learning; Qualitative/ordinal decision-making; Multiobjective/multicriteria decision-making; Preference learning/elicitation.
I have also a page on the DESIR website, which you can visit here.
Room 406 (26-00)
4 Place Jussieu
75005 Paris, France
Tel: +33 (0)1 44 27 88 64
Fax: +33 (0)1 44 27 88 89