Re: learnable dynamics (was Goal of AI...)

"JGCASEY" <jgkjcasey@xxxxxxxxxxxx> wrote:
Michael Olea wrote:
JGCASEY wrote:

I recognize x[i+1] = k*x[i]*(1-x[i]) from reading I have
done on chaos but have never actually looked much into
the subject.

The variable x is a line segment running from 0 to 1.

Thus in BASIC I guess you would do something like this,

k = 3
for x = 0 to 1 step 0.1
PRINT k * x * ( 1 - x)
next i

More like this:

k = 4.0
x = 0.4
LOOP:
PRINT x
x = k * x * (1 - x)
GOTO LOOP

So with your unsupervised learning algorithms do
they need a critic or do they just work by themselves
to learn a compact set of rules that are a "good"
approximation of the actual rules?

I believe Mike is simply talking about code that attempts to correctly
predict each number in the sequence. You don't normally think about it
having a critic but you could think about it as having a critic which
calculates the difference between the predicted value and each actual value
and uses that as a punishment (aka adjustment to the prediction system).
Most likely, any code that actually attempts to predict an infinite series
on the fly (one pass) will probably be working that way (or something
similar such as using the square of the differences as the punishment
value).

And do you think this is related to how the brain might
learn about its input? Is our input chaotic in any way?

Conceptually, very much so. Implementation wise, maybe, or maybe not.

--
Curt Welch http://CurtWelch.Com/
curt@xxxxxxxx http://NewsReader.Com/
.

Relevant Pages

  • Re: Sorting the posers from the pros
    ... search for the guilty ... punishment of the innocent ...
    (sci.military.naval)
  • Re: Evolution Befuddlement
    ... have gas, electric chair, poison injection, firing squad and did have ... How can John have 'predicted' beheading as a form of punishment? ... have been a prediction worthy of comment! ...
    (talk.origins)
  • Re: determinism, freewill, chaos, and circular causality
    ... fundamentally chaotic or wheter it's just too complex for practical ... To look for the basis of "free will" from chaos is like ... their own internal "self-behavior", ...
    (comp.ai.philosophy)
  • Re: Sorting the posers from the pros
    ... search for the guilty ... punishment of the innocent ...
    (sci.military.naval)
  • Re: Airlock door closes
    ... I'm not so sure about populations and chaos - I reckon there might be ... Chaos doesn't apply to people, ... To paraphrase Henry Ford: "Psychohistory is bunk". ... Psychohistory was never about prediction about individuals. ...
    (uk.rec.sheds)