Research On Bayesian Conditional Decision Theory

Conditional decision theory is an applied theory of probability logic that uses probability logic as the theoretical basis.It chooses to treat probability as a subjective interpretation of a person's belief in the proposition to assign a probability to the decision it uses.Bayes' theorem can infer the probability of an ambiguous event by the probability of an event associated with the basic properties of the event.Since conditional decision theory is based on such a Bayesian theorem,he can help decision makers make the best choices when conditions are not clear.In general,the main concern of this theory is which choice is the optimal solution for the decision maker.Conditional decision theory derives its name from the use of conditional sentences to calculate the expected utility of an action.Starting from solving the problem of "division invariance",Joyce used the imaginary method to process conditional sentences and used Jeffrey's view of expectations to establish conditional decision theory.One contribution of conditional decision theory is to construct a framework structure that combines evidence decision theory with causal decision theory.Conditional decision theory is a normative decision theory.It tells people from the perspective of logic and reasoning how to make the right decision,rather than describing what the correct decision is.Conditional Bayesian constructs a representation theorem for causal decision theory,which eliminates the difference in the form of evidence decision theory and causal decision theory,and embodies their common internal forms.However,the conflict between causal decision theory and evidence decision theory is still not solved,but it is transformed into the difference of state disjunction interpretation,and the basis of interpretation is derived from people's different intuitive understanding of reason.Judging from the development status of conditional decision theory,there are still many problems to be solved in the future.For example,for group decision problems,subjective probabilities are used instead of subjective probabilities.Construct a consistent Bayesian model for situations where multiple decision makers exist.In addition,we can consider how to expand decision makers from the category of tool rational people into a more realistic form;in addition,how to introduce other research tools in Bayesian conditional decision,such as preference logic,is also a problem to be solved.