Le 13-août-06, à 12:57, Stathis Papaioannou a écrit :
>
>
>
>
>
> Bruno Marchal writes:
>
>>>> I know it looks counterintuitive, but an AI can know which computer
>>>> is
>>>> running and how many they are. It is a consequence of comp, and the
>>>> UDA
>>>> shows why. The answer is:
>>>> the computer which is running are the relative universal number
>>>> which
>>>> exist in arithmetical platonia (arithmetical truth is already a
>>>> universal video game, if you want, and it is the simplest). How many
>>>> are they? 2^aleph_zero.
>>>> I have already explain it here:
>>>> http://www.mail-archive.com/everything-list.domain.name.hidden/msg05272.html
>>>>
>>>> It is a key point and we can come back on it if you have some
>>>> difficulties.
>>>
>>> Well now I'm confused! I thought the whole point of the earlier part
>>> of the UDA
>>> as discussed in the cited post (and many others of yours) is that you
>>> *can't*
>>> know the details of your implementation, such as what type of
>>> computer
>>> you are
>>> being run on, how fast it is running, if there are arbitrary delays
>>> in
>>> the program,
>>> and so on. Are you now saying that if I am being run on the 3rd of
>>> 100
>>> PC's in
>>> the basement of the local university computer science department, but
>>> everyone
>>> is keeping this a secret from me, there is a way I can figure out
>>> what's going on
>>> all by myself?!
>>
>>
>>
>> Did you read my old post to Brett Hall:
>>> http://www.mail-archive.com/everything-list.domain.name.hidden/msg05272.html
>> Perhaps you could comment it and tell me what does not convince you.
>> It
>> is indeed correct that the earlier parts of the UD Argument show that
>> a
>> machine cannot know about comp-delays, or about the real/virtual
>> nature
>> of a computer which would support the machine's computation until ...
>> you realize that for exactly those reasons the machine first person
>> expectations can only be computed (exactly and in principle) through a
>> measure on all possible computations (reducing physics to searching
>> such a measure). But then the first person can no more be associated
>> with *any* particular computations. Read the Brett Hall post where I
>> explain more, again in a "steppy" fashion (but the point is different
>> from the UDA).
>> To sum up that point:
>> 1) comp shows we cannot know which computations supported us.
>> 2) digging deeper in comp, this means eventually we are supported by
>> *all* (relative) computations (relative to some actual state; the
>> "actuality" itself is handled in the traditional indexical way, and
>> eventually "indexicality" is treated through the logic of
>> self-reference (G and G*).
>
> I did read the cited post. I can see that given step 7 - from the
> inside, we
> cannot tell if we are in a physical world emulation or a Platonia
> emulation -
> we could say that we are emulated by all appropriate computations.
OK. Even in a sort of "simultaneous" way. In a *probabilistic* sense we
are in all those appropriate computations. Like I said to Peter, even
if we are in a material universe (if that means something), then one
instant later we are not (with probability 1 minus epsilon).
> [I think
> it is still logically possible that we are emulated at any instant by
> one
> computation in the ensemble, and that there might still be a separate
> physical world, at this point at least.]
Yes. It is logically possible. But arithmetically or probabilistically
quasi-improbable.
> I can see that step 8 is right:
> if we are simulated in a massive real world computer, to make a
> prediction
> about the future we must take into account all the computations passing
> through our present state and define a measure on them.
All right, but then you should see that physics (let us see it as a
science of concrete and exact prediction) is already given by a purely
mathematical measure on a mathematical set (the set of computations
going through my actual comp state).
> I will accept step
> 9 on your authority, although this seems amazing: we can derive the
> laws of
> physics from a measure on first person computational histories.
I guess you mean that you don't see how to make the extraction. UDA
just shows that IF comp is true, then physics has to be derivable from
computer science/number theory.
To see how we can do the extraction is another matter. I do it by
asking UDA, no more to *you*, but to an universal machine.
> Now, step 10
> is my problem: if the laws of physics as per step 9 turn out to match
> our
> experimentally verifiable reality, then we are no more simulated by
> the physical
> computer than by any of the non-physical ones in Platonia.
>
> Why does it mean that we are no more simulated by the physical computer
> than one of the non-physical ones if the laws of physics match the
> prediction
> from the assumption that we are being simulated? Surely all that it
> suggests is
> that we are indeed being simulated - somewhere.
I was using the weak form of OCCAM. With comp we must always take into
account all simulations.
>
> You also state in step 10 that "...to say we belong to the massive
> computer has
> no real meaning: if it stops, nothing can happen to "us" for
> example...". That is
> true, but it does not mean that we are not actually being simulated in
> a
> particular computer, just that we cannot *know* which computer it is
> unless
> we have external knowledge. But in any case, this is just what I was
> saying
> before: you can't know which computer you are being simulated on.
OK. I should have said that we can know in which type of computing
machine we are. The "real one" which emerge from all computation, of a
fake one, just relatively material, because we are imprisonned by some
"higher level entities". Now if it is a fake one, we can in principle
find it (unless the entities kill us once we discover the fakeness of
the environment (like beieng lucid in a dream).
> Are you
> disagreeing with this statement because you believe you *can* narrow
> it down
> to knowing that you are not being simulated on a physical computer?
Yes, because, although I cannot know which computer implements me, I
can verify if me and my neighborhood is well described by the
comp-physics or by "artificial (fake) physics". If my neighborhood is
described by the comp-physics, then I know that the "computer" is just
the natural one associated to the platonic numbers with a probability
1-epsilon. If not, it means (like in a dream) that I am one level above
that reality, and I can bet on entities playing me some trick (like my
brain during the night).
We can never know we are NOT dreaming. This does not entail we can
never know we are dreaming. Dream. Awakeness is not symmetrical. OK?
Must go now. I will read the remaining posts tomorrow, and I will also
try to send a "very short roadmap",
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 everything-list.domain.name.hidden
To unsubscribe from this group, send email to everything-list-unsubscribe.domain.name.hidden
For more options, visit this group at
http://groups.google.com/group/everything-list
-~----------~----~----~----~------~----~------~--~---
Received on Mon Aug 14 2006 - 11:24:13 PDT