Possibilistic Preference Elicitation by Minimax Regret

Loïc Adam (1) , Sébastien Destercke (1, 2)

1 Heudiasyc - Heuristique et Diagnostic des Systèmes Complexes [Compiègne]
2 CNRS - Centre National de la Recherche Scientifique

Identifying the preferences of a given user through elicitation is a central part of multi-criteria decision aid (MCDA) or preference learning tasks. Two classical ways to perform this elicitation is to use either a robust or a Bayesian approach. However, both have their shortcoming: the robust approach has strong guarantees through very strong hypotheses, but cannot integrate uncertain information. While the Bayesian approach can integrate uncertainties, but sacrifices the previous guarantees and asks for stronger model assumptions. In this paper, we propose and test a method based on possibility theory, which keeps the guarantees of the robust approach without needing its strong hypotheses. Among other things, we show that it can detect user errors as well as model misspecification.

Mots clés: preferences elicitation; incremental; minimax regret; uncertainty; possibility theory

Domaines: Intelligence artificielle [cs.AI]

Loïc Adam, Sébastien Destercke. Possibilistic Preference Elicitation by Minimax Regret. 37th Conference on Uncertainty in Artificial Intelligence (UAI 2021), Jul 2021, Online, United States. pp.718-727. ⟨hal-03521623⟩