You don’t need a whole universe sim (especially if things outside your light cone don’t enter into your reward, which they shouldn’t!?).
-
-
This Tweet is unavailable.
-
This Tweet is unavailable.
-
Replying to @chc40
Uncertainty over these distant data are handled by summing over all those possible states. Which clarifies why the compute is so bad.
1 reply 0 retweets 0 likes -
This Tweet is unavailable.
-
This Tweet is unavailable.
-
Replying to @chc40
Gotta see a movie, but I’ll tag in
@MoralOfStory, who is good at this sorta stuff. Or that Sun can tag the right one if I misremembered.1 reply 0 retweets 0 likes -
Replying to @othercriteria @chc40
IMO the initial state of the universe has to be smaller than a human brain or we are all Boltzmann brains.
1 reply 0 retweets 0 likes -
This Tweet is unavailable.
-
Replying to @chc40 @othercriteria
It is a constraint on the size of the number. With infinite computation one can reproduce the universe to arbitrary precision given initial conditions and a predicate which returns true when run over the location of interest.
1 reply 0 retweets 0 likes
So Solomonoff probably never needs to consider a program longer than e.g. 10^15 (however big you think a human brain is) bytes to predict anything we'll ever see in this universe.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.