Re: Exploiting limitations of Turing machines in Turing tests?



In article <fdt045$k4c$1@xxxxxxxxxxxxx>,
Tero Hakala <tero.hakala@xxxxxx> wrote:
In one thing that TM's and brains seem to be different is that the
number possible of TM's is infinite, but countable (in the sense
that natural numbers are). While in working human brains, the
neurons get their information in analog manner from other neurons
(ie. real numbers come into play here) if we discount
quantization effects. So it seems that brains (or dynamics of brain)
are uncountable infinite and therefore exceed the number
of possible TM's.

This is true only if the "analog manner" you speak about cannot be simulated
by high-precision rational numbers. It is not at all clear that if one
perturbs the inputs to a neuron by, say, 10^(-1000000) then its behavior
will change.

And if we take the quantum mechanics into
account, we seem to get into a bottomless swamp. (And TM's, being
deterministic, would have also some difficulty to handle quantum
aspects.. as far as QM is considered valid in the form as it is now)

As far as computation is concerned, it is again not clear whether
quantum mechanics buys you anything. Any *finite* sequence of "truly
random" digits can be handled by the TM model of computation. An infinite
sequence cannot, but again it is not clear that such an infinite amount
of information can be "used" in the operation of the brain.

Can this be considered a valid distinction between brains and TM's?

Possibly, but it's not at all clear. The sticking point is that the
differences you posit rely on infinitely long sequences of numbers, or
an infinite amount of information, and it's not clear how such things
can be relevant physically. It's also not clear how we can test such
a theory in the lab, given that any experiment we can actually carry
out must take a finite amount of time and work with finite precision.

--
Tim Chow tchow-at-alum-dot-mit-dot-edu
The range of our projectiles---even ... the artillery---however great, will
never exceed four of those miles of which as many thousand separate us from
the center of the earth. ---Galileo, Dialogues Concerning Two New Sciences
.



Relevant Pages

  • Re: Review of Mueckenheims book.
    ... ....2222, they are actually infinitely distant elements of a sequence, ... smallest positive number, on the infinite scale. ... That's the ordering used by Cantor to prove their countability. ... If a set S is countable, then there is a total order < of S such that ...
    (sci.math)
  • Re: Well Ordering the Reals
    ... I have always allowed that sequences can be countably infinite, ... like the sequence of standard naturals is. ... TO nor anyone else has produced any axiom system in which it can be ... at several removes from pure mathematics, ...
    (sci.math)
  • Re: Chex Wat: Pi is "random" and "not predictable"?
    ... their output sequence. ... The fact is that an algorithm can also be used to ... first was about the probability of finding an apparent match. ... They may be matched within e at an infinite ...
    (talk.origins)
  • Re: Multiple infinities - one more look
    ... continued for lager length of digit sequences without limit. ... infinite digit sequences... ... so the resulting reals have an order. ... (i.e. having a finite program to output their digits in sequence). ...
    (sci.math)
  • Re: Calculus XOR Probability
    ... distances that make it a curve instead of a straight line, ... In the above sequence, n is a (strictly ... the diagonal is the "infinite case" of the staircases. ... closer to 2*pi as n -> oo. ...
    (sci.math)