On 03 Aug 2011, at 21:58, meekerdb wrote:
On 8/3/2011 11:13 AM, Bruno Marchal wrote:
I don't take it for granted. But I can imagine building an
intelligent robot that acts in every way like a person. And I
know that I could replace his computer brain for a different one,
built with different materials and using different physics, that
computed the same programs without changing its behavior. Now you
deny that this robot is conscious because its brain isn't made of
proteins and water and neurons - but I could replace part of the
computer with a computer made of some protein and water and some
neurons; which according to you would then make the robot
conscious. This seems to me to be an unjustified inference. If
it acts conscious with the wet brain and it acted the same before,
with the computer chip brain, then I infer that it was probably
conscious before.
Do I conclude that it experiences consciousness exactly as I do?
No, I think that it might depend on how its programming is
implement, e.g. LISP might produce different experience than
FORTRAN or whether there are asynchronous hardware modules. I'm
not sure how Bruno's theory applies to this since he looks at the
problem from a level where all computation is equivalent modulo
Church-Turing.
This is highly ambiguous. Obviously comp does not make all
computation equivalent. What happens in my head is not equivalent
of what happens in your head. Comp is just the statement that there
is a level where I am digitally emulable.
Church thesis is used only for making "digitally emulable"
mathematically defininable.
Then it is proved that comp implies that below my substitution
level *all* universal machine competes, and there is a big set of
equivalent computations (in the sense that I could not distinguish
them from a first person perspective).
I think you conclude that any Lobian machine is conscious (at some
level).
OK.
So what would be your guess, would two Mars Rovers programmed for
identical behavior, but one running a LISP program and the other a
JAVA program have the same conscious experience?
If by "identical behavior" you mean "identical behavior whatever the
inputs are", they will have exactly the same conscious experience
(when put in exactly the same environment). I suppose JAVA and LISP
manage well the (relative) real time constraints.
If by "identical behavior" you mean "identical behavior on Mars, such
precise day, such precise year", then it can depend on other factors.
Once you are simulated below your substitution level, it does not
matter if you are run by JAVA or by LISP. If the level is the neural
level, all what matter is that the right neural computation is
correctly emulated.
That is why below our level of substitution our bodies/matter have to
be a sum on the works of infinitely many universal computations. That
is why matter has to look quantum like, once we assume comp.
Would it make a difference if one ran on AMD and one on Intel? I'd
guess not.
You are right ... by default. There is an implicit engineering
assumption that Intel and AMD are not buggy, and can both satisfy the
local relative real time constraints. I can say "yes" to a doctor who
propose me a von Neumann computer for my brain, with only one
processor, but in that case, to stay in touch with the local reality,
it will need a superfast clock to emulate the asynchronous behavior of
my neurons and glial cells. That is not a problem for the UD, given
that it dovetails (super-slowly) on all relative computation, and that
is only the relative synchronicity and a-synchronicity which counts.
But I think it would make a difference if one used several chips
running asynchronously and the other used only a one CPU, or if one
had seismic sensors and the other IR sensors.
It can only mean that you have not copied the system, including the
environment, at the right level. See my answer just above. OK?
The artificial brain given by the doctor is always supposed to be able
to run your brain in "real time". If the artificial brain does not do
that, you will feel like the world around you is speeding-up (or
slowing down). You still survive from your point of view, but with a
relatively slower (or quicker) brain, and this will be an handicap in
your real life. But this is a red herring if it is used for stopping
the consequences of the comp. hyp. At the seventh step, the "doctor"
is replaced by a concrete UD in the universe, and at step eight, the
concrete UD in the universe is replaced by arithmetic.
I often prefer to consider "dreaming brain" to avoid this engineering
problem of interfacing a brain with an environment. The UD, in fine,
interface you with all possible local environment, executed by all
universal interpreters. What counts is the relative speed of the
computations. The UD will run, without doubt, the Heisenberg complex
rational matrix of the Milky-Way in LISP, and in JAVA. You cannot and
will not see the difference, and those two computations belong to your
<here and now> domain of first person indeterminacy, among many other
computations.
Bruno
http://iridia.ulb.ac.be/~marchal/
--
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.
