From: Bon Mo (bym198@ecs.soton.ac.uk)
Date: Thu Mar 08 2001 - 10:37:53 GMT
Re: Lucas: Minds, Machines and Goedel
> Clark:
> The type of randomising device suggested by Lucas is true random
> number generation - the number of radium atoms to have radioactively
> decayed in a container over a given period of time. This, unlike the
> standard pseudo-random number generation, will make the machine
> completely non-deterministic - there will be no way of predicting
> which option it will take when faced with a choice.
Mo:
Sometimes when making a choice, you do not want it to be completely
random. When humans need to choose, they assign weights to all the
options. The more correct facts that you know and believe in, that you
can associate with an option, the more probable choice the option becomes.
You implicitly assign probabilities of choice to each option. The
randomness should come in when all the choices have equal probabilities.
> Clark:
> Brains are basically causal systems - something happens, and neurons
> fire; this is turn creates another effect, such as movement - they
> therefore can be considered to be "mechanical".
Mo:
A causal system can be physical or probabilistic, but a important property
of these systems is that we understand their mechanisms. In the case of
the brain we do not understand its mechanism, therefore we cannot declare
it a causal system.
> Clark:
> if we were to build a mind-modelling machine that was
> Turing-indistinguishable from a real mind, and started to take pieces
> off it, there would come a point where the machine would no longer be
> intelligent. Therefore, the same must be true when working in the
> opposite direction.
Mo:
Where would this intelligence performance have a boundary? What components
of a human brain would a mind-modeling machine be required to implement
to be assigned "Turing-indistinguishable" and intelligent?
This archive was generated by hypermail 2.1.4 : Tue Sep 24 2002 - 18:37:21 BST