Re: my current position (was: AUDA)

[Marchal]

Wei Dai wrote:


Suppose someone offered you $1000, but if you accepted Earth
would be destroyed and everyone on it killed as soon as you die. Would you
take that offer? Even if you did I'm sure most people wouldn't.


This is because I include in the first person its possible
compassion feeling for what are possible others. (This is 
similar to what Brent Meeker said in its last post).
Compassion, although it bears on others, is a feeling, isn't it?


I admit the decision theory approach I gave in the last post has problems,
some of which you've pointed out. But what's the alternative? I've been
thinking about this issue for several years, starting with the
expected-first-person-experiences approach (if you read the earlier
archives you'll see many posts from me on this).


I did. (and it is why I try to understand your evolution).
The very beginning of this discussion-list by you and Hal Finney has been
very attractive to me at the start. I still don't understand your shift.


GTM means general Turing machine. It's defined in Jürgen
Schmidhuber's paper at http://www.idsia.ch/~juergen/toesv2/. Please read
it if you haven't already.


I red it. I prefer its first paper, although its philosophical
conclusions contradict what I found interesting in it. We discussed that.


Wait, even in the infinite the ratios will not be the same in general. Why
do you think they will be?


Not so easy question indeed. But here the methodology I use forces me
to define the measure by the AUDA logic Z1*. The infinite ratios will be
the same thanks (hopefully) to the non trivial constraints given by
computational self-reference. Remember that our first person expectations
rely on *all* our consistent (self)-extensions.


If you read Schmidhuber's paper, you'll see that he offers several
measures for consideration. He believes that the Speed Prior is the
correct objective measure,


This cannot be. The UDA shows our first person experiences cannot be
aware of delays taken by any universal (classical or quantum, but
immaterial) machine accessing our current states. Classical real time
is definitely an emerging phenomena from UD* (all execution of UD).
UDA predict that we are (perhaps) quick to be computed but our
neighborhoods must be necessary much slow to be computed. (In that
sense it predicts the computing power superiority of our neighborhood).
You can interpret my work as saying that IF we are made of bits then 
we are necessarily embedded in a qubit made reality.
I have made some recent progress in that direction. Normally Z1* should
be equivalent to some sort of generic quantum computer.
The incredible progress in that field could lead more quickly than
I expected to a refutation or confirmation of comp.


Suppose you want to crack a bank's encryption key, which is worth $4
million to you, and there are two ways to do it. You can spend $2 million
to build a quantum computer to crack the key, or you can spend $3 million
to build a classical computer to do this. Now if you believe the Speed
Prior is the correct measure, then you'll think that the quantum computer
will very likely fail, and therefore you should go with the classical
computer instead.But if you believe the Universal Prior is the correct
measure, then you'll think that both computers will work and you'll go
with the quantum computer because it's cheaper.


OK.


However, there's another way to think about this situation that doesn't
involve an objective measure. The fast-to-compute and the slow-to-compute
universes both exist.


You are taking the expression universe too literaly. The slow-to-
compute multiverse is equivalent to the sheaf of locally quick to
compute single computations, but we belong to the mutiverse: we belong
to all universes. Objective measure are useful for taking into account
the proportion of histories and this is what makes decision worthly
senseful.


(The fast-to-compute universes are the ones where
quantum computers fail.) So when you adopt the Speed Prior you're really
saying I know the slow-to-compute universes exist (and my actions affect
what happens in them), but I just don't care very much about those
universe.


But (sorry for repetition) the UDA forces us to take those slow
universe/computation into account. That's exactly the point of question 
7 in the conversation with Joel Dobrzelewski.
(links at http://www.escribe.com/science/theory/m3044.html, step
7 is at http://www.escribe.com/science/theory/m2992.html).
You must care about those slow universe because their slowness
just comes from the fact that their multiply you in important
proportion. It is the very base of my proof that comp entails the quantum,
and why if we are bit-describable then those bit are qubit made.


To me the attraction of think about it the second way is that it allows us
to just say that all universes exist. We don't have to say that
objectively one universe has a higher measure than another. What does that
mean anyway? If all universes exist, 

Re: Finite time and infinite space

[Matthieu Walraet]

On 15 Jan 2002, at 11:31, [EMAIL PROTECTED] wrote:

 One of the things that strikes me as most peculiar and unexpected about the
 universe is this: that it is apparently finite and inhomogeneous in time,
 yet infinite and homogeneous in space.
 

The universe is finite :
My short term memory last less than 10 minutes. So the universe is 
contained in a sphere of 600 light-seconds radius around my head.

Matthieu.
-- 
http://matthieu.walraet.free.fr

Re: Kiln People

[hal]

Wei writes:
 This brings up the question: Which measure is evolution making us try to
 maximize? The answer is none. It only appears that way because people who
 try to maximize their measures according to some measure function will
 tend to have large measures according to that measure function. So if
 you sample the multiverse according to some measure function,
 you'll likely find people who appear to be trying to maximize their
 measures according to that measure function. But if you then sample the
 multiverse according to a second measure function, you'll likely find
 people who appear to be trying to maximize their measures according to the
 second measure function.

This makes sense in a formal, mathematical way.  Given a sheaf of
universes you can apply any weighting function you want.  But I still
think you are taking too many degrees of freedom here.  My intuition is
that there must be some kind of constraint which keeps you from adopting
arbitrary measures.  But I don't have a good idea yet for what that
might be.

A couple of possibilities occur to me.  One is that it might be
irrational to adopt other measures (for some definition of rationality).
For example, rationality might impose some consistency conditions on your
weighting function.

Another possibility is that mathematics says that there is really only
one measure function, the universal measure, for all but an insignificant
fraction of worlds.  That is, all measure functions are arbitrarily
close to the universal measure, in the limit.  I thought I remembered
reading that this was a property of the universal measure.  If so then
it would mean that you can't really depart from it very much.

There is also the point I and others have made, that you are not just an
observer from outside the universe, but a participant inside.  This ties
you to the universe in a way which might constrain you.  I think your
argument above makes more sense if you think of yourself as an observer
of a multiverse in which you are not participating, say some kind of
computer simulation.  Then the idea of measure seems pretty arbitrary.
However once you are inside you are influenced by the reality of measure.
I don't see how you can reconcile the notion that measure is arbitrary
with the observation that the laws of probability work.  Aren't these
phenomena tied together?  Living here in this world, aren't you forced
to either believe that the universe is fantastically improbable (because
we live in an extremely low-measure universe for some arbitrary measure),
or else that we do in fact live in a high measure universe, meaning that
measure is not arbitrary?

I think you have answered this last objection already but I need to think
about it some more.  I don't know if any of these proposals really work.

Hal

Re: my current position (was: AUDA)

[Brent Meeker]

Hello Wei

On 15-Jan-02, Wei Dai wrote:
 On Tue, Jan 15, 2002 at 12:47:18PM +0100, Marchal wrote:
 This is because I include in the first person its possible
 compassion feeling for what are possible others. (This is 
 similar to what Brent Meeker said in its last post).
 Compassion, although it bears on others, is a feeling, isn't it?
 
 But what is the compassion about? In this case it's about events that
 you'll never experience in the first person. If you want to reason
 about your compassion and make rational decisions based on it, you
 have to do it from the third-person point of view.

I don't understand reason about your compassion.  The point is that
you have a feeling about a possible future you imagine and so you take
action to avoid that future.  It doesn't have to be something after you
die.  If you decide not to stroll across the freeway because you have a
bad feeling about getting hit by a truck and so you decide not to
stroll across the freeway you have made a here-and-now decision about
avoiding something - and hence never experiencing it.  You are
reasoning about your actions with your feeling (compassion?) as part of
a premise.  You're not reasoning *about* your feeling (which I suppose
would be called '3rd person').  

Brent Meeker
  If people are good only because they fear punishment and hope for
reward, then we are a sorry lot indeed.
-- Einstein