“…We mention in particular [1] which explores the relation between probability and nondeterminism, and [3] which extends of the notion of probable innocence. A related line of work is directed at exploring the application of information-theoretic concepts [24,12,20,21,29,5,10,19,15,8,9,17,18,2]. The relation with hypothesis testing is given by the fact that the exponential of the conditional entropy is an upper bound of the Bayes risk (the probability of error using the MAP rule) [23,4], although [26] has pointed out that the bound can be in some case very loose.…”