Abstract

Decision theory does not traditionally include uncertainty over utility functions. We argue that the a person’s utility value for a given outcome can be treated as we treat other domain attributes: as a random variable with a density function over its possible values. We show that we can apply statistical density estimation techniques to learn such a density function from a database of partially elicited utility functions. In particular, we define a Bayesian learning framework for this problem, assuming the distribution over utilities is a mixture of Gaussians, where the mixture components represent statistically coherent subpopulations. We can also extend our techniques to the problem of discovering generalized additivity structure in the utility functions in the population. We define a Bayesian model selection criterion for utility function structure and a search procedure over structures. The factorization of the utilities in the learned model, and the generalization obtained from density estimation, allows us to provide robust estimates of utilities using a significantly smaller number of utility elicitation questions. We experiment with our technique on synthetic utility data and on a real database of utility functions in the domain of prenatal diagnosis. 1

Citations

6234	Maximum likelihood from incomplete data via the EM algorithm - Dempster, Laird, et al. - 1977
990	Decision with multiple objectives: preferences and value trade-offs - Keeney, Raiffa - 1976
543	Linear Algebra and Its Applications - Strang - 1988
512	Optimal Statistical Decisions - DeGroot - 1970
416	Bayesian classification (AutoClass): Theory and results - Cheeseman, Stutz - 1996
293	Influence diagrams - Howard, Matheson - 2005
247	A tutorial on learning Bayesian networks - Heckerman - 1995
155	Efficient approximations for the marginal likelihood of Bayesian networks with hidden variables - Chickering, Heckerman - 1997
101	Graphical models for preference and utility - Bacchus, Grove - 1995
93	Learning Gaussian networks - Geiger, Heckerman - 1994
92	From influence diagrams to junction trees - Jensen, Jensen, et al. - 1994
91	Dynamic programming and influence diagrams - Tatman, Shachter - 1990
82	Making rational decisions using adaptive utility elicitation - Chajewska, Koller, et al. - 2000
6	Can preference scores for discrete states be used to derive preference scores for an entire path of events? Medical Decision Making 17 - Kuppermann, Shiboski, et al. - 1997
3	The Beaver Dam Health Outcomes Study: initial catalog of health-state quality factors,”Medical Decision Making - Fryback, Dasbach, et al. - 1993
1	Associations between health status and utilities implications for policy - Lenert, Treadwell, et al. - 1999