A fractal CPU

[David Makin]

For anyone into sci-fi who'd like to read a good book loosely related to this thread try the "World of Tiers" series by Philip Jose Farmer or "Time enough for love" by Robert Heinlein.

[kram1032]

The latter part of that is arguable if talking about the whole of existence since such an entity is itself part of existence...i.e. the unanswerable question "Where did God come from then ?" or "Who created God ?".

This is correct. It's not scientifically answerable. Which is part of the reason why I'd rather keep these subjects apart.

[Apophyster]

I'm trying to not bounce off too many walls.  My apologies for the ot diversion.
@kram1032: I struggled to find a better word than "purposed", maybe it's "reached".
If I may add, I tend to ignore discussions about brain since it's our mind that *we* use and are familiar with.
And I ain't got no business at all talkin' philosophy. 

@David M.: I liked also "The Moon is a Harsh Mistress" and before that "Stranger in s Strange Land".
I totally admire all your fractal work as well.
I think I shall retired before the last wall I try to bounce from cracks my skull and I shall no longer have to worry over brains, minds, thoughts, mentations, ideations, cognitions, and the various other troubling matters that occur between my ears.

Fred E

[kram1032]

But I didn't have troubles with the word "purposed" - it rather was the word "seek" that didn't work well here...

[jehovajah]

Good survey kram, of the empirical  basis of our knowledge so far. The paradigm one uses is crucial to the explanatory platform one uses. Thus my use of the fractal principles to frame these Questions. and to organise   the empirical data. The models then constructed have some sympathy with any individual experience. The search for elegance becomes understandable, as does the complexity.

Your last post shows the futility of words, for semantics leads to that feeling, where shared experience allows us to pull together.

[kram1032]

Hmmm..
Calling words useless seems a bit far of a stretch.
That we even have words is one of the most important reasons for a medium like the one we're using right now, to even exist.
Written language isn't always sufficient to give away emotions and such but it's excellent for delivering core information of statments from one end to another.
However, if what you were trying to say is that words can be misinterpreted so easily then, yes, that's entirely true.

Having a scalefree view of the world is imensely powerful. It's the paradigm I choose to frame most of my perceived world in aswell. There almost is no point in life where it simply doesn't apply and usually, when it does not, it's simply because you lack some particular information that obviously links a situation to a way broader set of situations.

However, how about we try to come back to the actual topic. Or is everything said and done for that already?
Fractal CPU - refering to the imperfections of our brain.
I was pointing out that those imperfections are a result of two things:
- it's "too hard" as in it takes too many additional resources to make the process more accurate
- a certain level of inaccuracy can actually be benefical for survival in that it spawns creativity and new ideas.

Also I was pointing you to a couple of resources of which I'd especially recommend one:
Human Behavioural Biology - lectured on Stanford University.
It covers many patterns of human behavior on all scales: That professor organizes his lectures like that: He starts with what happens "right before" you behave in a certain way, e.g. a certain signal arives at a certain part of the brain, then he goes back in time further and furter, usually covering how that behavior depended on your life when your mother was pregnant with you - this is, where epigenetics strongly comes in - and finally he goes all the way back to how it happened in evolution.

By doing this overview, he is really amazing at giving you the whole, scale-free picture.

[jehovajah]

Yes, very good.
I value some words personally for their role in my knowledge accessing structures. Some other words i have to create myself to access the insight i have. But words are tools for personal use more than for communication. It is by shared experience that we may use words to communicate.
Scale free is nice, but having worked so hard to reintroduce the fractal paradigm, i pesonally am not willing to move away from the word fractal, even though the word may move away from me!
Your succinct summary of your long post is wonderful for moving back to the thread topic. I will reread it again with these directions in mind!

[kram1032]

Really, the difference between scalefree and fractal is mostly the difference between continuous and discrete.
For instance, in the Koch kurve won't look qualitiatively entirely the same if you don't zoom in at least to a certain level

But a Wiener-process, actually looking very similar to that (Wiener-processes are a generic concept of which, for instance, brownian motion is a part) will look much smoother while zooming in. There is no visible step-size for repetitions to occur.

The difference is quite subtle but it's there.
However, since it's so subtle, you can almost use those terms interchangably.

[jehovajah]

So Lar sayst current computers try to avoid error causing structures, but the biological brain does not. Lar then reveals that his religious choice is rompting tis thread, and that what i have called the biological brain has an intelligent design brief, basically to accumulate useful knowledge about what seems a random world of swirling chaotic events.
He then presupposes that god is the simplest answer to any problem or question and that the human /biological brain will arrive at this conclusion.

He then states yjat theoretically a computer ith a fractal chip design should come to the same conclusion.
He then asks for a discussion on his Belief which he has put in a hypothetical form, and thus called a theory. The questionseems to be, build these systems and test if my hypothesis is correct? Which is not a question but a recruitment of fellow thinkers.

Kram comes along with the current scientific and evolutionary based, still forming consensus view , basically letting us know that there is some real empirical research evidence and speculation we can imbibe before or while we discuss lar2's theory, Basically if we want a proper and philosophical discussion on this topic of the biological brain lets all start on the same page!

lar2's thread is not based on deep empirical data, but on deep and resilient faith. A discussion may proceed if the parties declare their aims.

The reason i joined the thread was because i just had another thought about how the biological brain and the electromechanical processing chip would be virtually identical could the chip be fractalised to the same extent. I do not see the need to include or exclude god or gods in this. Kram however alludes to a powerful systematic connectedness in every human process through the fractal biological structure, which is itself completely explicable by fractal, generative, iterative processes such as evolution. i am in sympathy with that view and systematic explanatory force, but again do not see the need to exclude or include gods in the discussion.
For if you believe god or gods have done this , then you should marvel at your gods; or if you believe evolutionary processes have done this then you should marvel at their fractal patterns; and so on.

However, if it is your intention to convert one believer to your belief, then declare it. Otherwise let us have sensible discussion and speculations about the empirical data as far as we know it.

If i have not fairly stated your views please correct me.

[kram1032]

Hmm, this is a very interesting way to put it all. To summarize your summary, you agree with me on rather looking at it from an iterative point of view aswell as that you feel, discussing this doesn't require (as in prove or disprove) any form of higher entitiy/-ies.
That seems correct to me. I think, you captured quite well my take on the topic.

However, I have the feeling that you didn't capture Lar's take on it quite as well. I can't quite put my finger on the reason for that though. Most of it seems correct but *something* seems off. Maybe it's the conversion thing. I don't think, he tries to force anyone into any kinds of opinions.
But we'll see. Hopefully, Lar will, as you put it, "declare his aims".

I'd like to add that many physicists and mathematicians nowadays are astonished or blown away by how simple, beautiful, elegant many complex phenomenas' descriptions wind up to be and some of them will tell you that it's hard to NOT believe in some kind of higher beeing's plan when these solutions emerge.

I've even seen a t-shirt somewhere stating:

Because those four simple equations essentially define all there is to say about electromagnetism and offsprings such as optics.

[jehovajah]

The tee shirt says what i tried to express succinctly. The  bigger sentence contains the equations. We can just discuss the equations, and leave the larger frame to individual choice.

I not only think a fractal processor would mimic human signal experience in the brain, even more so if electrolytes can be involved, but other biological brains are similar to the human signal experience. Animates differ by degree not by quality of processing.

[kram1032]

There indeed are only very few differences between our brains and those of fruit-flies, if we're constraining our scope to the building plans of the invovled neurons.
The main differences are that our brains are significantly more sectioned and more voluminous.
As I *think* I already have mentioned, this sectioning goes into an extreme in authistic people. That clearly shows how significant this sectioning is to behavior. Authistic people are often extremely adept logicians, so they are, in a way, much more like a computer than the average person.
But they have a very hard time living. They loose a lot of their emotional capabilities (that doesn't mean they are unable to express any kinds of emotions but it sure becomes a lot harder for them) and in fact, to even understand an opposing individual's emotions, they have to focus quite a bit and really think about what the external body language tells them about it. (I'm pretty sure though, as always, there isn't a pure black and white - how much this is the case certainly depends on the Individual) - there are no implicit clues they are picking up, or almost none. They essentially have to learn about emotions like a computer might be able to with a machine-learning algorithm. (That being said, they are still a lot better in that than any computer nowadays could be)

This simple example should show quite clearly, why it's not a good idea to have perfect logics in a brain. It comes with a penality of error-robustness and general survivability.
In fact, the most important part of our brain, subject to survival, is also the most primitive one: The stem-brain is almost exactly like it already existed in reptiles. It encodes the most important aspects of survival, like running from pretadors, running towards prey, breathing, changing heart-frequency...
Have a damage there and you're pretty much dead.
Have a damage in your frontal cortex and you can't stick to any rules anymore - e.g. your social skills and your morale totally fail. Up to the point where you can entirely state a rule in a very precise and fully correct form but you can't for the love of whatever apply it.
BUT. You survive. (Until people get so mad at you that they execute you. But that's a different story)
There are excellent examples of this in that lecture about Human Behavioral Biology from Stanford University. I know I repeat myself but I really recommend you all to watch that.

[krispychicken]

I think more important than the structure of this computer is how is receives information. Ours uses the senses, or experience. How would this computer experience? User inputs, algorithms? How exactly does our brain store experience? If it is indeed an iterative structure, would if form itself? These are essential questions.

[kram1032]

Ah, you undug THIS topic again. Nice and timely, because I have found something related:
Check out numenta grok.
It seems to have entered commercial phase recently, but it's essentially an implementation of a so called "HTM"-network - a Hierarchical temporal memory.
The inventor of them gave a google tech talk (and various other talks but they all are fairly similar and just vary in length and detail)
<a href="http://www.youtube.com/v/4y43qwS8fl4&rel=1&fs=1&hd=1" target="_blank">http://www.youtube.com/v/4y43qwS8fl4&rel=1&fs=1&hd=1</a>

It's essentially mimicing as closely as possible by means of a computer, a small region of the neo-cortex - notably, pretty much ANY small region of it, because, as it turns out, it's roughly the same everywhere.

This is the closest to a fast implementation of a biological brain we've ever had.
(There are really slow implementations that do model way more details, but they simply aren't feasible for actual realtime work)
They are still working on improving this, modelling more details, but this is essentially as close as we are.

He also gives great insight of how the brain's memory and information processing works.
Check out the talk and if you want *slightly* different angles on it, search for more of them.

There also is an open implementation here:
http://sourceforge.net/p/openhtm/wiki/Home/
It's in C# and thus only works on Windows so far, but there is a guy who tries to port it to C, I think, with the goal of being multiplatform in mind.

[John Smith]

Goodness, I need to look at my post replies more often.  I thought this thread had been unraveled a long time ago.  I assure you that, reading this discussion, I do thoroughly regret mentioning God.  I should have realized everyone on this forum has wide ranging beliefs on that highly sensitive subject.  My point is: our brains, and any animalian brain, for that matter, succeed where man-made computers largely fail.  The brain has many qualities that make it unique, my question is which of these qualities should we imitate if we wish to build computers with similar capabilities.  Obviously one could just culture brain cells inside the cpu. That has actually been tried.  But few people can afford to keep a normal computer running, let alone a biological one.  I wanted to know what quality you could imitate most cost effectively to produce the same or similar results.

@kram1032:  Love the T-shirt.  If any one on this forum thinks I really believe God stepped out of wherever he lives and said, in English, "Let there be light!" or in any other human language for that matter, I find it no wonder he finds my beliefs ludicrous.