Utilities as Random Variables: Density Estimation and Structure Discovery

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]  Daphne Koller,et al.  Making Rational Decisions Using Adaptive Utility Elicitation , 2000, AAAI/IAAI.

[2]  M. Kuppermann,et al.  Can Preference Scores for Discrete States Be Used to Derive Preference Scores for an Entire Path of Events? , 1997, Medical decision making : an international journal of the Society for Medical Decision Making.

[3]  L A Lenert,et al.  Associations between health status and utilities implications for policy. , 1999, Medical care.

[4]  R. Klein,et al.  The Beaver Dam Health Outcomes study , 1993, Medical decision making : an international journal of the Society for Medical Decision Making.

[5]  M. Degroot Optimal Statistical Decisions , 1970 .

[6]  Frank Jensen,et al.  From Influence Diagrams to junction Trees , 1994, UAI.

[7]  John B. Kidd,et al.  Decisions with Multiple Objectives—Preferences and Value Tradeoffs , 1977 .

[8]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[9]  E. Rowland Theory of Games and Economic Behavior , 1946, Nature.

[10]  David Heckerman,et al.  Learning Gaussian Networks , 1994, UAI.

[11]  Fahiem Bacchus,et al.  Graphical models for preference and utility , 1995, UAI.

[12]  Audra E. Kosh,et al.  Linear Algebra and its Applications , 1992 .

[13]  Peter C. Cheeseman,et al.  Bayesian Classification (AutoClass): Theory and Results , 1996, Advances in Knowledge Discovery and Data Mining.

[14]  A. Rosenfeld,et al.  IEEE TRANSACTIONS ON SYSTEMS , MAN , AND CYBERNETICS , 2022 .

[15]  Ross D. Shachter,et al.  Dynamic programming and influence diagrams , 1990, IEEE Trans. Syst. Man Cybern..

[16]  F. B. Vernadat,et al.  Decisions with Multiple Objectives: Preferences and Value Tradeoffs , 1994 .

[17]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[18]  R. L. Keeney,et al.  Decisions with Multiple Objectives: Preferences and Value Trade-Offs , 1977, IEEE Transactions on Systems, Man, and Cybernetics.

[19]  Craig Boutilier,et al.  Context-Specific Independence in Bayesian Networks , 1996, UAI.

[20]  R L Kane,et al.  Methodology for measuring health-state preferences--I: Measurement strategies. , 1989, Journal of clinical epidemiology.