Re: Exploiting limitations of Turing machines in Turing tests?

Antti Valmari wrote:
Tero Hakala wrote:
In comp.ai Don Geddis <don@xxxxxxxxxx> wrote:

I was recently contemplating Turing tests and Turing machines (TM) and was
wondering if the fundamental limitations of TM can be exploited to discover
whether the conversation partner in a Turing test is a digital computer AI
or a real person.
No, they can't. Because humans are subject to the same limitations.
Can you elaborate this a little bit? I was not aware of any proof
that humans can't exceed computational power of TM's. While there
are arguments that the human thought process might be understood/
modelled by a computer analogy, there a lot of contrasting views also.

There is strong evidence to the claim that anything that can be computed
with a physically realistic computer, can be computed with a Turing
machine.

For one thing, many seemingly different computational models have been
suggested. Some of them have proven strictly weaker than Turing
machines, and most of them have the same computational power as Turing
machines. None of them is stronger than Turing machines, except those
which contain a physically unrealistic element to start with (like an
oracle that has immediate access to a well-defined but infinite
database). On the other hand, very modest machinery suffices for
obtaining Turing-strength. The class of computable functions is thus
very robust and insensitive to how it is defined. Google Church--Turing
thesis.

All those models do incorporate a limitation that does not apply to
human thinking - inability to produce a wrong answer.

People often face situations in which failure to make a decision is
certain to lead to a bad outcome, so it is better to decide, even if the
decision is wrong. When facing a saber-toothed tiger, it may be better
to jump either left or right than to stand still thinking about which
way to jump.

I think our brains have evolved to make some decision, even if it may be
wrong.

Consider the halting problem. Programmers often say a computation is in
an infinite loop even without having done the analysis to distinguish an
infinite loop from a very bad performance problem in a computation that
would ultimately terminate.

Maybe probably approximately correct machine learning is a better model
than basic computability?

Patricia
.

Relevant Pages

  • Re: Turing Machines and Physical Computation & TM S REVISIONISTS
    ... >> Postivism implies some sort of physicalism, I think, or at least it ... I do not condone those idealist re-readings of Turing ... > THAT!), THEN, it is absurd to talk of ANY MACHINES WITH INFINITE SIZE. ...
    (comp.theory)
  • Re: Turing Machines and Physical Computation & TM S REVISIONISTS
    ... >> Postivism implies some sort of physicalism, I think, or at least it ... I do not condone those idealist re-readings of Turing ... > THAT!), THEN, it is absurd to talk of ANY MACHINES WITH INFINITE SIZE. ...
    (sci.math)
  • Re: Language chatter
    ... our "practical computing machines" are Von Neumann machines rather than ... And Von Neumann machines derive from a model of a human ... I don't think the actual unification of Church and Turing models ... I'll certainly defer to the memory of a guy who was there;-). ...
    (comp.lang.ruby)
  • Re: Turing machines
    ... Turing machines are executing a program too, but it's a very special program. ... means that no computational device, no matter how complex or advanced, can do ... Except that many folks believe humans ...
    (comp.ai.philosophy)
  • Re: Disproof of the Halting Problems Conclusion
    ... > is specifically limited to Turing Machines. ... > states that the solution set of the Halting Problem is limited to Turing ... are equivalent in computing power. ...
    (sci.logic)