Elicitation of a Utility from Uncertainty Equivalent Without Standard Gambles
Labreuche, Christophe; Destercke, Sébastien; Mayag, Brice (2015), Elicitation of a Utility from Uncertainty Equivalent Without Standard Gambles, in Destercke, Sébastien; Denoeux, Thierry, Symbolic and Quantitative Approaches to Reasoning with Uncertainty.13th European Conference, ECSQARU 2015, Compiègne, France, July 15-17, 2015. Proceedings, Springer : Berlin Heidelberg, p. 25-35. 10.1007/978-3-319-20807-7_3
TypeCommunication / Conférence
External document linkhttps://hal.archives-ouvertes.fr/hal-01200696
Conference title13th European Conference, ECSQARU 2015
Book titleSymbolic and Quantitative Approaches to Reasoning with Uncertainty.13th European Conference, ECSQARU 2015, Compiègne, France, July 15-17, 2015. Proceedings
Book authorDestercke, Sébastien; Denoeux, Thierry
Number of pages554
MetadataShow full item record
Thales Research and Technology [Palaiseau]
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Abstract (EN)In the context of decision under uncertainty, standard gambles are classically used to elicit a utility function on a set X of consequences. The utility of an element x in X is derived from the probability p for which a gamble giving the best outcome in X with probability p and the worst outcome in X otherwise, is indifferent to getting x for sure. In many situations, uncertainty that can be observed on the true value of X concerns only neighbour values. Uncertainty is then represented by a probability distribution whose support is an interval. In this case, standard gambles are unrealistic for the decision maker. We consider uncertainty represented by an equi-probability over an interval of X. This paper addresses the elicitation of a utility function on X by obtaining the certainty equivalent of an equi-probability over an interval of X. We show that not all utility models are suitable to accomplish this task.
Subjects / KeywordsIntelligence artificielle
Showing items related by title and author.