wedrifid comments on Skill: The Map is Not the Territory - Less Wrong

49 Post author: Eliezer_Yudkowsky 06 October 2012 09:59AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (174)

You are viewing a single comment's thread. Show more comments above.

Comment author: wedrifid 09 October 2012 12:43:29PM 3 points [-]

I don't think it's possible to fail rationality "for ever"

Merely humanly impossible. If you are a more pure agent just assign probability "1" to enough things and you'll be set.

Comment author: CCC 10 October 2012 01:31:34PM 0 points [-]

Hmmm. It seems that I should add "as long as you are able to reassign all priors of 1 to priors of 0.999999999, and all priors of 0 to priors of 0.000000001" to my list of exceptions. (It won't fix the agent immediately, but it will place the agent in a situation of being able to fix itself, given sufficient observations and updates).

Comment author: Eugine_Nier 10 October 2012 10:57:45PM 2 points [-]

That's not the only problem. An agent that assigns equal probability to all possible experiences will never update.

Comment author: CCC 11 October 2012 07:07:27AM 1 point [-]

Oh, that's sneaky.

Perhaps a perfect agent should occasionally - very occasionally - perturb a random selection of its own priors by some very small factor (10^-10 or smaller) in order to avoid such a potential mathematical dead end?

Comment author: Eugine_Nier 12 October 2012 12:56:37AM 0 points [-]

Nice try, but random perturbations won't help here.

Comment author: CCC 12 October 2012 07:15:49AM 0 points [-]

I think that this re-emphasises the importance of good priors.