Reformulating decision theory using fuzzy set theory and Shafer's theory of evidence

摘要: Abstract Utilities and probabilities in decision theory are usually assessed by asking individuals to indicate their preferences between various uncertain choices. In this paper, we argue that (1) The utility of a consequence can be as the membership function fuzzy set ‘ satisfactory ’. (2) probability an event, instead being directly assessed, should inferred from evidence associated with event. degree is quantified using Shaferian basic assignments. addition, use Heisenberg Uncertainty Principle for change one technical assumptions underlying theory. As result change, some kinds will observable certain experiments but unobservable others. Since defined over potential outcomes experiment, they only some, not all, evidence. result, different could inconsistent. This formulation emphasizes importance new distinctions (and just information) updating probabilities. We addresses many observed empirical deviations experiment. It also anomalies quantum physics. close brief discussion directions further research.