On 27 Sep 2011, at 21:25, meekerdb wrote:
On 9/27/2011 1:27 AM, Bruno Marchal wrote:
On 26 Sep 2011, at 21:44, meekerdb wrote:
On 9/26/2011 9:08 AM, Bruno Marchal wrote:
Suppose that you are currently in state S (which exist by the
comp assumption).
But what does "you" refer to?
Your first person view. Or the owner of your first person view,
restricted to that view, without salvia amnesia, if you want.
The comp assumption seems ambiguous. Is it the assumption that
"you" are instantiated by a specific computation?
No. Something like that can be part of the consequence, but this is
clearly not assumed. In fact the UD shows that "you" is
instantiated by an infinity of computations.
Or is it the assumption that your brain could be replaced, without
you noticing, by a physically different computer, so long as it
computed the same function (at some level).
Yes.
These seem slightly different to me and are only identical if QM
is false and the world is strictly classical and deterministic.
At a practical level the brain is certainly mostly classical and
so I might say 'yes' to the doctor even though my artificial brain
will have slightly different behavoir because it has different
counterfactual quantum behavior. But this difference seems to
present a problem when trying to identify "you" within the
inifinite bundle of computations instantiating a particular state
in the UD computations.
Why? If my "original brain" is described by QM (without collapse)
it might be said to self-multiply naturally. But that self-
multiplication will be contagious on the UD in that universe, so
this will not change the relative proportion.
That's the step that seems ambiguous. What you write above applies
to a physically realized (i.e. quantum) UD, but not to the UD in
Platonia.
Why. In platonia the UD is multiplied, in all possible ways, including
the quantum one.
The physically realized UD will have non-zero probabilities of doing
something random instead of implementing the intended function.
And? What would that change anything in the reasoning?
And what do you mean by physically realized?
On the contrary, the UD itself forces a multiplication to be lived
from inside.
As to identify yourself in the UD*, this is just impossible in any
third person ways. But the indeterminacy is on the first person
experiences, not on their description in the UD. So the statistics
are lived from inside. A computation is winning, if indeed you feel
to be alive through its UD instantiation.
Ambiguities remain, but they are part of the measure problem.
Of course if you replace the whole universe with an emulation,
instead of just my brain, then my emulated brain in the emulated
universe can have the same behavior as my natural brain in this
universe.
Yes, and that is why the reasoning will work in the limiting case
where your "generalized brain" is the entire universe described at
some level. The UD will generate all the digital approximation of
that universe, and at some level of approximation, you will not see
the difference, because we are assuming comp.
The UD generates an infinity of computations going through that
state. All what I say is that your future is determined by all
those computations, and your self-referential abilities. If from
this you can prove that your future is more random than the one
observed, then you are beginning to refute rigorously comp. But
the math part shows that this is not easy to do. In fact the
random inputs confer stability for the programs which exploits
that randomness, and again, this is the case for some formulation
(à-la Feynman) of QM.
How is this?
Consider the iterated self-duplication experience, like with the
random movie, where you expect to see (correctly) a random movie.
The movie will seem random because the limiting case is described
by a Gaussian (accepting the p = 1/2 for a single duplication).
Other considerations make such a randomness occurring below you
substitution level, so it might be that the only way to stabilize
the computations above the substitution level comes from some phase
randomization, similar to Feynman explanation of why QM minimize
the path action.
So you're talking about keeping the computation classical, even
though realized by a physical device which is microscopically quantum?
Computation is a classical notion at the start. Also, the UD Argument
does not assume quantum mechanics, nor any physical theories. It
assumes only that a physical reality exist do that the notion of
doctor, hospital, brain, and enough concrete computer can exist, so
that it makes sense to say "yes" to a digitalist doctor.
At the start of the reasoning it is better to be agnostic on
physicalism.
I don't recognize the reference to "the random movie".
It is the thought experience where you are multiplied in 2^(16180 *
10000) versions, each in front of a different image on a black and
white screen having 16180 * 10000 pixels. And this 24 times per
second, during 90 minutes.
The question is: what is more probable among the following: you will
feel to be
- in front of a black screen for one hour and half
- in front of a white screen for one hour and half
- in front of a screen with the movie "Shadow of a Doubt "by Hitchcock
- in front of white noise snow, but actually it gives the first (16180
* 10000) * (60 * 90) * 24 decimal of Pi.
- in front of white noise
- - in front of a screen with the movie "Shadow of a Doubt "by
Hitchcock with chinese subtitle.
Now, note that the UD, by its excessive dumbness, multiplies each
computation, in that way, by dovetailing it on the real as dummy
argument, so that our comp global (in front of a UD) indeterminacy has
to take this into account.
We need a notion of negative (amplitude) of probability,
Negative probability or negative, imaginary probability amplitude?
We have good evidence for the last (imaginary probability amplitude),
but only the future will decide. Well, technically the Z and X
material hypostases are going in the same direction.
extracted from comp, for such a procedure to work, but this is
already provided by the logic of self-reference when we add the non-
cul-de-sac assumption (Dt) to the provability modality (Bp), with p
sigma_1. This can be made enough precise to make sense of how the
quantum can be explained by the digital viewed from the digital
creature themselves. No doubt that a lot of work remain to be done,
but that is exactly what I wanted to show.
You lost me.
Let me put it in this way. I interview classical platonist machines,
being ideally correct (they never emit arithmetically unsound
statements). Bp = the machine rationally believes p (and this means
that the machine believes this instinctively (p is some axiom), or can
derive it from such axioms by using the usual inference rules.
Knowledge is given by Bp & p (following Theaetetus).
Observation is given by Bp & Dt (so that we get the "probability"
notion in the way made obligatory by the UD Argument). Dt is the non
cul-de-sac assumption, so that "probability" has a meaning.
Physical observation is the same as observation, except that p has to
be restricted to the sigma_1 sentences, so as to restrict the
probabilities on the UD computations. This gives the Z1 and Z1* logic.
The logic of quanta observable appears in Z1*. The qualia appears in
X1* (the same but with the modal variant given by Bp & Dt & p; we just
reapply the Theaetetus idea).
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.
