I don’t understand. Does that mean they have priors?
I think it’s another way of putting it, though IIRC the biases are not always explicitly prior probabilities, they could just be a way the algorithm is constructed. Choosing the specific construct is acting on a prior.
I don’t understand. Does that mean they have priors?
I think it’s another way of putting it, though IIRC the biases are not always explicitly prior probabilities, they could just be a way the algorithm is constructed. Choosing the specific construct is acting on a prior.