My opinion is that quantum mechanics is essential to define an OM,
despite it being in the classical domain. The computational state of an
AI is not the precise physical state of the system that generates the
AI, it is some coarse grained picture of it. So, if you have a
classical computer, then the bits that are zero or one only become
visible when you average over the microstates.
Then, even the observer does not appear at the level of the bits, you
need to extract the information that is present in the bits, and there
must be a huge redundancy there too. What we are aware of are patterns
in the information that enters our brain, but the same pattern we're
aware of can be realized in an astronomically large number of ways.
Therefore, if you are aware of something right now, the exact quantum
state that describes this is, in general, an entangled state which
contains the correlations within the patterns that you are aware of and
the information present in the environment that are mapped to those
patterns.
This state defines the program your brain is running, at least as far
as rendering the patterns you are aware of.
Saibal
Citeren meekerdb <[email protected]>:
On 9/27/2011 3:55 PM, Jason Resch wrote:
On Sep 26, 2011, at 6:31 AM, Stathis Papaioannou <[email protected]> wrote:
On Mon, Sep 26, 2011 at 7:45 AM, meekerdb <[email protected]> wrote:
An interesting talk relevant to what constitutes an "observer moment".
http://www.youtube.com/watch?v=0VQ1KI_Jh1Q&NR=1
Even if the experience is smeared out over time
I think it is clear with mechanism that this is the case. Imagine
an AI with a single CPU. Here it is obvious that it's state extends
through the dimension of time. With the parallel processing of the
brain it is less, but still much greater than a Planck time.
Even assuming signals at c the brain extends about a nano-second in
time, 22 orders of magnitude longer than the Planck time.
But doesn't this create problems for Bruno's argument, which assumes
states are timeless, instant like things in Platonia and that they
have no overlap. Should we identify observer moments with bundles of
UD computations going thru the same state, but also with extensions
of those computations forward and backward over some number of
states? But they are not the same forward and backward. Or do we
require that the "substitution level" be pushed down to time slices
short compared to a nano-second so that an observer moment will be a
whole set of states extending over a short time. In which case the
sequence of states will pick out a much smaller set of UD
computations that went thru all those states.
Brent
and has a complex
relationship to real world events it could still be the case that it
can be cut up arbitrarily.
Perhaps arbitrarily in the sense of distinct observer moments, but I
don't think so about time.
There is no way I can be sure the world was
not created a microsecond ago
Consider how many CPU cycles are required for the AI to become
aware. Even if you think it becomes conscious as soon as the first
instruction is executed, the instruction takes some amount of time
to complete.
If it takes the brain 100 ms to compute a moment of awareness, then
you can know you were not created 1 microsecond ago.
Jason
and there is no way I can be sure there
isn't a million year gap between subjective seconds.
--
Stathis Papaioannou
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
