id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_asd3qlm4h5df5d22p6f65gqcnm	Bas C. van Fraassen	Updating Probability: Tracking Statistics as Criterion	2016	19	.pdf	application/pdf	7715	571	66	What is generally called Bayesian Conditionalization is a policy for updating probability prior probability function, and allows as possible posteriors the conditionalizations on If the prior probability assignment had the possibility that it was tracking the is that the input that triggers a change in probabilities may not be of the sort this policy response to a given input, to form a posterior probability function. The Bayesian policy, applicable here, starts with a prior probability, takes the inputs to probability function p (the 'prior') a set R of possible 'posteriors' (of cardinality greater relevant statistics, and updating the probability assignment on new input should preserve updating policy could allow for all the convex combinations of a given finite set as the set The probability measures form a convex subset of this vector space, defined by In such examples, where the input is a new value for a conditional probability of a given 	./cache/work_asd3qlm4h5df5d22p6f65gqcnm.pdf	./txt/work_asd3qlm4h5df5d22p6f65gqcnm.txt