Was this meant to be a reply to my bit about the Solmonoff prior?
If so, in the algorithmic information literature, they usually fix the unnormalizability stuff by talking about Prefix Turing machines. Which corresponds to only allowing TM descriptions that correspond to a valid Prefix Code.
But it is a good point that for steeper discounting rates, you don’t need to do that.
It was inspired by yours—when I read your post I remembered that there was this thing about Solomonoff induction that I was still confused about—though I wasn’t directly trying to answer your question so I made it its own thread.
Was this meant to be a reply to my bit about the Solmonoff prior?
If so, in the algorithmic information literature, they usually fix the unnormalizability stuff by talking about Prefix Turing machines. Which corresponds to only allowing TM descriptions that correspond to a valid Prefix Code.
But it is a good point that for steeper discounting rates, you don’t need to do that.
It was inspired by yours—when I read your post I remembered that there was this thing about Solomonoff induction that I was still confused about—though I wasn’t directly trying to answer your question so I made it its own thread.