I'm doing a thing with a Bayesian Network. There is a tool to analyze such networks and there is a "doubt" setting in [0, 1]. If the certainity of a prediction is less than that value, then it is classified as a doubt and not a prediction. OK, seems straightforward... But then there is a second setting called delta, and it works the same way only "the probability distribution of a random variable is itself a random variable" (pseudo-quote). (This is the only background I can give you as it doesn't say more.)
I am totally confused by that as I don't see how you even can separate those two terms: isn't a probability distribution of a random variable simply a description of how likely that variable is to take on each of its possible outcome values?
Any clarifications (or if you find som flaws in my reasoning) are welcome :)