Doesn't this depend on what is AIXI's reward function? If the reward function involves exactly predicting the bit that will emerge at time t, that's one thing. If the reward function involves predicting the distribution of a series of bits occurring over time, that's another thing...
-- Ben On Mon, Apr 28, 2014 at 3:48 PM, Alex Miller <[email protected]> wrote: > (Please let me know if this isn't the best forum for this kind of question) > > Suppose we give a long sample from the Bernoulli distribution to AIXI. > I would expect an intelligent agent to figure out that it cannot see > any patterns there and be able to continue the sequence accordingly. > > However, I think the shortest programs that reproduce such a sequence > (the ones that should dominate the hypothesis space, according to > AIXI) be the ones that copy the given sequence and then produce > something highly regular, e.g. all zeros. Is this correct? > > > ------------------------------------------- > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/212726-deec6279 > Modify Your Subscription: https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com -- Ben Goertzel, PhD http://goertzel.org "In an insane world, the sane man must appear to be insane". -- Capt. James T. Kirk "Emancipate yourself from mental slavery / None but ourselves can free our minds" -- Robert Nesta Marley ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
