Research On Bayesian Causal Decision Theory

Bayesian decision theory is based on causal logic of probability, which is to say this theory is an application of probabilistic logic. Bayesian probability interpretation of the probability calculus is a collection of individuals with regard to the reasonableness of the rules of personal confidence or behavior associated with confidence or consistency conditions.As a technical support of causal bayesian decision theory, bayes' theorem means that when one can not predict the probability of an event, he can rely on the probability of an event which is essentially related the former. Therefore, bayesian decision theory is the causal theory for decision makers in case of uncertain information making judgments and choices. The core of the theory is to find the best choice for policy-makers. Different from classical and evidential of decision theory, causal decision theory is to calculate conditional probabilities using virtual expected utility of action, which is the reason of the name of the causal decision theory. Because the probability of virtual condition should be traced back to the causal relationship, it will make the theory of causality, when using them to calculate the expected utility of an action. The feature of Bayesian causal decision theory mainly reflects in the action efficacy causality and probabilistic causality.Basically, the causal bayesian decision theory is a normative theory rather than descriptive theory. It suggests people how to make a decision logically,rather than describe what people actually do. According to Bayesian causal decision theory and probabilistic logic, Bayesian network is established which shows the method of Bayesian applied in the field of artificial intelligence.From the developing trend of the causal bayesian decision theory, it can be studied in the future. For example, the main subjects are to study the multi subject problem, to consider whether the beliefs and values of multiple Bayesian decision makers can be merged into one, and to represent their common preference for selection. In addition, we can also consider the causal bayesian decision problem outside the scope of instrumental reason and the introduction of preference logic research tools to examine the causal bayesian decision problem.