Dear Wei,
Interleaving.
----- Original Message -----
From: "Wei Dai" <[EMAIL PROTECTED]>
To: "Stephen Paul King" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Monday, December 23, 2002 5:16 PM
Subject: Re: Quantum Probability and Decision Theory
> On Wed, Dec 18, 2002 at 08:54:30PM -0500, Stephen Paul King wrote:
> > Two ideas would seem to mute this strange thought.
> >
> > 1) The no-cloning theorem, iff the world follows QM and not just
classical
> > physics.
>
> Are you saying the no-cloning theorem will prevent copying of minds?
[SPK]
Yes. I strongly suspect that "minds" are quantum mechanical. My
arguement is at this point very hand waving, but it seems to me that if
minds are purely classical when it would not be difficult for us to imagine,
i.e. compute, what it is like to "be a bat" or any other classical mind. I
see this as implied by the ideas involved in Turing Machines and other
"Universal" classical computational systems.
The no cloning theoren of QM seems to have the "right flavor" to explain
how it is that we can not have first person experience of each other's
minds, whereas the UTM model seems to strongly imply that I should be able
to know exactly what you are thinking. In the words of Sherlock Holmes, this
is a "the dog did not bark" scenario.
> What about AIs running on classical computers?
>
[SPK]
It would help us to find out if an AI, running on a classical computer,
could pass the Turing test. We seem to have become so extreme in our
rejection of Lucas and Serle that we have fallen off the edge and assume
that zombies could not exist.
> > 2) van Fraassen's Reflection Principle seems to ignore the possibility
that
> > probabilities and their distributions are given ab inition and that the
> > notion of "updating" and/or revising one's expectations is negligible.
>
> I don't understand what you're trying to say. Can you expand on it please?
[SPK]
I wrote incorrectly. What I wrote should have read:
> > 2) van Fraassen's Reflection Principle seems to ignore the possibility
that
> > probabilities and their distributions are NOT given ab inition and that
the
> > notion of "updating" and/or revising one's expectations is negligible.
Let me see if I can find the paper that I read about the Reflection
Principle and see if I can be a bit more pointed in my argument.
See:
http://www.brown.edu/Departments/Philosophy/homepages/weatherson/Dutchmar.pd
f
In it we find:
***
Van Fraassen on Reflection
In Belief and the Will (1984) van Fraasen argues that we should follow a
principle called Reflection. If we
think that tomorrow we'll believe that the probability of p is x then today
we should believe the probability
of p is x.
***
It was this passage that lead me to the belief that Van Fraasen assumed
that the equality p is x is one that is not subject to updating. In other
words that our expectations based on current information are not subject to
revisions such that tomorrow's p is x will be identical with today's.
What I see implicit in this discussion is an attempt to deal
quantitatively with the notion of expectations. I am reminded of having read
somewhere about games theories where the palyers did not have complete
knowledge of the other player's possible moves. Brian Weatherson's paper,
from which I quoted, seems to imply such an idea but does not address it
directly.
Any comments?
Kindest regards,
Stephen
