Magnus,

Thanks for your effort to explain this. I appreciate the difficulty. I am
left with many questions but the bottom line is "Bring it On". Who would
have thought that Moore's Law would hold as long as it has? If we can jump
start a whole new level I'm for it.

I went to a presentation by Intel at Comdex several years ago. Bill Maher
was the host and he was asking a panel of experts why computers were such a
pain. He said something about buying a computer five years ago and now it is
all obsolete and what a drag that was. He picked the wrong audience for that
crack. The audience booed and the panel members, including Craig Barrett,
CEO of Intel informed him that the reason his old computer is obsolete is
that he can now buy one twice as fast. This is a good thing.

BTW, Maher's comment that computer geeks love Vegas because they find out
there is another use for silicon, went over much better with the crowd.

I guess I need to looking into the QC business a bit further for technical
details but it seems one of your main points has gotten lost. At least in
principle it is easy to imagine computers that are powerful enough to be
considered sentient. This has been explored at some length in sci-fi, from
Commander Data to Marvin, the HHGG robot with a Real People Personality.

Computers are already augmenting human intelligence to an astonishing
degree. Hardly a discussion in this forum takes place that doesn't send me
scurrying to Google to look up something I have forgotten or to check out
something I never heard of.

Those with an anthropist bent or the MIMOAT crowd are going to look for any
way they can to stay on top of the evolutionary game of King of the Hill.
They remind me of the bunch who claimed "Man the Tool User" was unique in
the world. Then we found chimps and even otters making and using tools. So,
"Man the Speaker" and Possessor of Language was King for a Day until we
taught a few chimps to use symbol systems and honeybee engaging in modern
dance. Kind of gets down to " the Sole Owner of the Opposable Thumb"... Now
that is special.

It is not so much the facts or principles that are at stake in whether there
are or can be machine intelligence. It is pure chauvinism.  After all there
are already bits of code out there displaying life-like properties. Viruses
spread and reproduce themselves across the internet everyday. I even saw a
headline in the Weekly World News (the last bastion of great American
journalism) that computer viruses are being spread to humans... This was
followed by an update on the World's Fattest Cat eating the World's Fattest
Mouse and Satan's face appearing in the skies over Washington.

Case







-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Magnus Berg
Sent: Saturday, February 17, 2007 4:20 PM
To: [email protected]
Subject: Re: [MD] Quantum computing

Hi Case

Thanks for your questions about quantum computers. They are probably a good
way 
to start explaining what quantum computers are about.

> [Case]
> Quantum computing is a bit new to me but I am having trouble seeing how
> allowing interference of any sort into the calculations can produce a
happy
> result. If I am look for a solution to a problem in cryptography how is
> evolution going to help? On the one hand it is suggested that qbits will
> allow much faster calculations and on the other that the processing power
> stems from quantum effects. Are these quantum effects resulting in raw
speed
> or in better Quality answers?

First of all, I *don't* have an in depth understanding of exactly how
quantum 
computers work, nor a deep knowledge in quantum theory. But I've read pretty

much about it so I think I have some grasp of what it's about. You can
probably 
find better answers online but here's my take on it.

Quantum Computers (QC from now on), is partly about raw speed, but they are
not 
fast in GHz. The speed comes from massive parallelism. And we're not just 
talking 4, 8, 16 or even 80 cores (as Intel claims to do in 2012). We're
talking 
2^n simultaneous calculations, where n is the number of qubits in the
system. 
So, for the 16 qubit QC just demoed, that amounts to 65536 simultaneous 
calculations. The funny, and crucial, thing to understand here is that these

calculations are really SIMULTANEOUS. If you don't believe that you might as

well stop reading, because that's just one thing that makes QCs conceptually

different from digital computers (DC). Some researchers claim the QC uses 
parallel universes to make the calculations, but how it works is not really
the 
point. The point is that this is *one* way that QCs are really different
from DCs.

However, to extract the result from a QC, you need to detect the states of
the 
qubits. This is done using some kind of QC/DC interface and will collapse
the 
superpositions of the qubits into digital states, a 0 or 1. And here's
another 
very important thing to realize with regards to the MoQ, when the QC/DC 
interface collapses the superposition of the qubits into a 1 or 0, it's a 
Quality Event. Such an event is very similar to what happens when I'm typing

right now. The qubits in my brain collapses/decides what to write and then 
writes it on my computer.

>>From what I could find about the demo you mentioned it looked promising
but
> there are skeptics and I don't suppose I will be looking at super-cooled
> chips on my desktop anytime soon. But then I would never have though I
would
> have a dual core processor on my desktop by now either so who knows. 
> 
> I wonder sometimes how appropriate computer metaphors are when thinking of
> biological systems. Truly they are useful but how far can we push them. We
> have been quick to seize the latest in technology and apply it to our
> understanding of biological processes. In the early days of telephony the
> brain could be seen as a complicated switching system, the brain is a
> thinking machine etc. 
> 
> Much of the cognitive sciences proceeds from the notion the brain is a
> sophisticated computer running programs. But this metaphor can break down
> when pushed. A program is an algorithm, a series of sequential steps. I am
> not sure what organisms do is at all sequentially. Computers operate in
> discrete steps or clock cycles I don't think organisms work this way.
> Certainly computer metaphors are useful but it is easy to overlook the
> limitations. 

To some, I will probably sound like someone reinventing the wheel, but I
really 
think QCs are conceptually different from today's DCs. The metaphors (and
many 
limitations) of DCs doesn't apply to QCs. To claim that the brain is like a
QC 
*is* very different from claiming it's like a DC.


> What modern efforts in AI are doing is attempting to simulate in
algorithmic
> form what brains do organically. So far it seems to me that the value of
> this accrues mainly from the effort involved in thinking this through.
That
> is we know the brain does pattern recognition; how can we construct an
> algorithm to recognize patterns. Digitally this requires a brute force
> approach.

Exactly, but if (when) they are able to use a QC for pattern recognition, it

will completely change the playing field. Think neural networks. A DC
handling a 
neural network must simulate every transition in the network, and every
added 
node in the network doubles the time it takes to calculate a result.

But a QC is a natural fit for a neural network. The brain *is* a neural
network, 
and a QC *is* a neural network. Adding a new node is "just" a matter of
adding a 
qubit. It will take *no* additional time to calculate a result. The problem 
being pondered at the moment is how to add an arbitrary number of qubits,
and 
even if that seems very hard, I'm sure they will start to increase from the 
current 16 in a decade or two.


> When you suggest that the brain maybe a quantum computer, how so? Aren't
> these attempts at quantum computing still handling binary information?
> Aren't they still going to be handling information in discrete steps?

Not internally, but as I wrote above, if you want to see a result from the
QC, 
you need to perform the QC/DC conversion step, the Quality event.

> I am
> not sure how quantum effects would apply to a nervous system. It is my
> understanding that quantum effect disappear above the level of complex
> molecules. Are you suggesting that macro level information is somehow
> processed at the level of protein synthesis?

Actually, quantum effects *can* affect larger structures. I'm not sure how
they 
did it in the demo QC but one way is to use a Bose Einstein condensate.  As
far 
as I remember, a BEC structure can be created in a superconducting
structure, 
which is probably why they need such powerful cooling. (Actually, I think I
read 
they use photon traps for the demo QC, and then they use entangled particles
to 
detect the states).

But if you're asking how quantum effects can affect a nervous system, like a

brain, I think BECs can explain that as well. Dana Zohar wrote in "The
Quantum 
Self" that there are some experimental results indicating that BEC
structures 
are affecting the inner working of the brain. But I'm on pretty thin ice
here, 
so don't take my word for it, at least not before I can re-check the source.

Best regards,

        Magnus


moq_discuss mailing list
Listinfo, Unsubscribing etc.
http://lists.moqtalk.org/listinfo.cgi/moq_discuss-moqtalk.org
Archives:
http://lists.moqtalk.org/pipermail/moq_discuss-moqtalk.org/
http://moq.org.uk/pipermail/moq_discuss_archive/

moq_discuss mailing list
Listinfo, Unsubscribing etc.
http://lists.moqtalk.org/listinfo.cgi/moq_discuss-moqtalk.org
Archives:
http://lists.moqtalk.org/pipermail/moq_discuss-moqtalk.org/
http://moq.org.uk/pipermail/moq_discuss_archive/

Reply via email to