I’m not sure this is actually more computable than the Solomonoff prior. Now, instead of having to do a potentially infinite amount of work to find the shortest program that behaves in a particular way, you have to do a definitely infinite amount of work to find all the programs that behave in a particular way.
Could you give an example of an inference problem that’s made easier by doing this?
I’m not sure this is actually more computable than the Solomonoff prior. Now, instead of having to do a potentially infinite amount of work to find the shortest program that behaves in a particular way, you have to do a definitely infinite amount of work to find all the programs that behave in a particular way.
Could you give an example of an inference problem that’s made easier by doing this?