Hi John,
On 14 Jan 2009, at 14:53, John Mikes wrote:
> Stathis: (from reply to Brent):
> "I return to my question about what would happen if there were a
> discontinuity in a sequence of states,..."
> There IS discontinuity if the state "transits'(?) from s1 to s2.
> Do you have any idea how one can observe a changed state by only
> continuous transitions? Where do you pick the limit to call it the
> 'next' state? Isn't such 'arbitrary'? (I have similar questions with
> astronomy - what I never studied - how do they calculate the
> planatary movement markers (year?) into (fractional?) seconds? but
> they do).
> Same question to Bruno's 'integers: there is no transition between 2
> and 3.
But I can offer you a machine transforming 2 to 3, and so on, and one
is enough for usual standard third person point of view on numbers.
Now, oversimplifying myself for reason of shortness: when you take
into account the personal interview of 2 and 3; what they infer
normally or generally is that there have to take into account an
infinity of computations going from 2 to 3. There are many transitions
possible, perhaps too many.
> Even putting fractals in between only extends the question to
> smaller scale.
Right.
> "Infinite small" I would not consider at our present level of
> cognition.
Me too. But, concerning the substitution level, which non constructive
existence is assumed as working hypothesis, we have to conceive the
idea of arbitrary small. The Mandelbrot set offers a vivid
illustration. Any plane compact representation of the universal
deployment will look like that:
People can take a look
http://www.youtube.com/watch?v=8cX6Ghis6ts
> ----------------------------
> In your reply to me:
> I find the rat-related IBM publication wanting:
> "...Each of its microchips has been programmed to act just like a
> real neuron in a real brain...."
> How much we know about a 'real neuron' and its function is a matter
> of the present level of R&D. We know more than a decade ago and less
> than a decade hence. And the 'rat' discloses only its movements -
> evaluated and understood(?) at the complexity of the human brain.
> Would you draw valid conclusions on - say - personality by a silent
> film of only the movements of a person - even at matching complexity?
> I may write some distracting stories to follow 'movements'.
> A rat doesn't communicate with researcher.
How can know that? And are you sure that the researcher you have in
mind is listening to the rat?
If you have the time look at those youtube videos, they provide a
counterexample.
Here a rat communicates something to a cat:
http://www.youtube.com/watch?v=7ikm3o5hDks&feature=channel_page
Here a cat communicates something to a human:
http://www.youtube.com/watch?v=TZ860P4iTaM&feature=related
:)
> Rat-shrink?
> I wouldn't use Terry Seynowski's critical word that the brain is too
> 'mystrerious': it is too complex and poorly followable by our
> present level of our 2009 cognitive inventory.
Agreed. But Lobian Machines can prove theorems on Machines: they can
prove that all Lobian machines are too complex to "understood" itself.
All lobian machine are too complex from its own point of view.
Yet some Lobian machine A can prove first that all correct machine
have about the same abstract theology, and second that they can study
the complete theology of some simpler correct Lobian machine. This
include the physics of the machine (physics is one of the
"theological" hypostases) so they can compare the structure of their
apparent neighborhoods with the machine's physics, and eventually even
measure some possible degree of non-computationalism.
> We can know just 'that' much and most likely there is much 'more' to
> it. (my 'enrichment' remark of past and future knowledge).
> Markram's work is glorious. That's the way we can proceed in
> widening our knowledge 10,000 'neurons' is a good start.
> But their fundamental tenet:
> "Every brain is made of the same basic parts...." -
> means a restriction to our ongoing physical/physiological
> observational capabilities which have shown incredible enlargement
> in the past (still within the 'physical world' figment).
> Frequencies, methods, evaluations are all limited.
> I keep it open that penetrating the Hard Problem we may find new
> phenomena unassignable to our present knowledge of the known tissue
> and physics. Nobody has diversified amp or MRI data to distinguish
> whether a blood-surge refers to a political opinion, love, boredom,
> financial expectation, or whatever, in a mentally-topical distinction.
> *
> Thanks for that brilliant article. It says honestly:
> "When listening to Markram speculate, it's easy to forget that the
> Blue Brain simulation is still just a single circuit, confined
> within a silent supercomputer. The machine is not yet alive. "
> Exactly.
> Neuroscience is heroic and 'explains' lots of tenets into open
> "somehow'-s. Their assumptions are within our conventional sciences.
> Digitally computed.
> Computing analogue ideational meanings?
Science works only with assumptions. And science practice works only
with bets.
I am open to any theories, but comp is cute because it gives startling
information. It relies on a big discover: the mathematical notion of
universal machine which I find awesome at many levels in all
directions. I think the digital makes things less prone to
reductionism, even if I can acknowledge the apparent contrary, but
then the Lobian machine can explain the origin of that apparent
contrary.
The universal machine can already demolish *any* theory about its
behavior that *you* could present to it. It is an eventual universal
dissident, made to live on the border in between security and freedom
(which is again well illustrated by the Mandelbrot set border).
Best,
Bruno
> On Wed, Jan 14, 2009 at 4:40 AM, Stathis Papaioannou <stathisp.domain.name.hidden
> > wrote:
>
> 2009/1/14 John Mikes <jamikes.domain.name.hidden>:
> > Stathis,
> >
> > common sense, not always applicable to math-related topics
> > is startled before a task on a REGULAR contraption-type Turing
> machine
> > (binary, electrically driven finite hardware etc.) can emulate ALL
> the
> > potentials of 11+billion neurons in unrestricted groupings and
> unlimited
> > connectivities as to the complexity of all the codes/details
> > (un!)imaginable.
> > (Maybe if you change to Bruno's infinite Loebian vs. Turing
> machine...?? I
> > doubt if you can do that, since there are different brains (eg for
> genetical
> > etc. reasons) and I cannot figure so many (although limited number)
> > variables in the 'unrestricted' (all encompassing?) Loebian
> machines.)
> It is possible to calculate how much computing power it would take to
> simulate a brain at a particular level. For simulations at the
> cellular level, there is for example this work by IBM researchers
> simulating a rat neocortical column:
>
> http://www.seedmagazine.com/news/2008/03/out_of_the_blue.php?page=all&p=y
> http://www.research.ibm.com/journal/rd/521/djurfeldt.pdf
>
> It's still a long way from simulating an entire brain and observing
> ratlike behaviour, but it does show that computational neuroscience is
> now beyond the philosopher's thought experiment stage.
>
>
>
> --
> Stathis Papaioannou
>
>
> >
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?hl=en
-~----------~----~----~----~------~----~------~--~---
Received on Wed Jan 14 2009 - 14:09:33 PST