Artificial brain 'ten years away'

@richrf,

richrf;79622 wrote:

Building tinker toys does not equate to replicating the human brain.

These are Tinker Toys:

http://www.charm.net/%7Ejriley/TinkerToy1.JPG

This is what Blue Brain is based on:

http://tekniikanihmelapsi.com/wp-content/uploads/2008/06/bluegene.jpg

Blue Gene - Wikipedia, the free encyclopedia
[INDENT]Blue Gene is a computer architecture project designed to produce several supercomputers, designed to reach operating speeds in the PFLOPS (petaFLOPS) range, and currently reaching sustained speeds of nearly 500 TFLOPS (teraFLOPS). It is a cooperative project among IBM (particularly IBM Rochester MN, and the Thomas J. Watson Research Center), the Lawrence Livermore National Laboratory, the United States Department of Energy (which is partially funding the project), and academia. There are four Blue Gene projects in development: Blue Gene/L, Blue Gene/C, Blue Gene/P, and Blue Gene/Q.[/INDENT](It's kind of a big deal.)

If you referring to those pix I posted, then yeah maybe eye replacements count as Tinker Toys, too.

richrf;79622 wrote:

I don't hold it against you if you can get some research grants.

You seem to think research grants are spent on caviar and cocaine.

(Well, somebody might need to study sturgeon, but that's not my department.)

richrf;79622 wrote:

Hopefully more innocuous than building stuff that can spy on people

I'm interested in that too

http://www.spectrum.ieee.org/images/public_html/automaton/MEMS08_BTI2.jpg

Cyborgification of moths, lol, reminds me of that song Stalker by Covenant: "In transit you pass among the strangers of the world, paying tribute to the thief who stole away your shadow"

richrf;79622 wrote:

but my own preference would be to spend the money on growing nutritious foods and building parks for children to play in.

Rich

Have we already descended into "world's smallest violin" territory

---------- Post added 07-26-2009 at 01:56 PM ----------

xris;79633 wrote:

It appears that opposing scientific prophesies is evoking magical quotations, im a witch, a wizard condemned for being sceptical.

This is a favorite tactic of magicalists: tu quoque

1 Reply

@odenskrigare,

Nannna..oh dont be so childish.If you cant debate without resorting to silly pictures and playing the superior scientist,dont bother.Why are you so offended ? all i have said is dont believe this guys claim to have sex driven imitation brain in ten years.Put it in a desirable body, i might just contribute to its developement.

1 Reply

@xris,

xris;79642 wrote:

all i have said is dont believe this guys claim to have sex driven imitation brain in ten years..

Why

You haven't given a reason

Is your expertise greater?

1 Reply

@odenskrigare,

odenskrigare;79643 wrote:

Why

You haven't given a reason

Is your expertise greater?

I dont believe a machine can have emotions ,a sex drive,what will it desire?an innate desire to survive?You can insert a programme to drive but what will its purpose be? you need to have a certain awareness,aware that it is a machine and then what?.You will invent a brain damaged child with no abilities beyond feeling certain cravings.A human brain is more than a reptilian shi----ing machine,will it say hello daddy?

1 Reply

@xris,

xris;79648 wrote:

I dont believe a machine can have emotions ,a sex drive,what will it desire?an innate desire to survive?You can insert a programme to drive but what will its purpose be? you need to have a certain awareness,aware that it is a machine and then what?.You will invent a brain damaged child with no abilities beyond feeling certain cravings.A human brain is more than a reptilian shi----ing machine,will it say hello daddy?

Um

In case you haven't noticed, our brains are electrochemical machines

1 Reply

@odenskrigare,

odenskrigare;79649 wrote:

Um

In case you haven't noticed, our brains are electrochemical machines

Your playing the elitist scientist again,why should i have not noticed?
Tell me its sex drive how will it manifest itself? mammalian is more basic than the reptilian and its desires are based on physical needs,how will you think this will manifest itself when it has no body to secure these desires?
I can make a dog salivate over a bone but it wont look at the stars.Lets see if they can make a worm and then make it desire a lettuce leaf before they make such outrageous claims.I look at what they have done and then judge them on what they think they can.

1 Reply

@xris,

xris;79651 wrote:

Your playing the elitist scientist again,why should i have not noticed?

So ... are you saying the brain isn't an electrochemical computer?

xris;79651 wrote:

Tell me its sex drive how will it manifest itself? mammalian is more basic than the reptilian and its desires are based on physical needs

Reptilian brains are universally less complicated than mammalian brains. In fact the reason the human brain looks so wrinkly and gross is because a relatively huge neocortex was wedged in with a bunch of other older parts, including that region of the brain we share in common with reptiles

xris;79651 wrote:

how will you think this will manifest itself when it has no body to secure these desires?

That's actually a somewhat valid point, and so what would be necessary is the replication of the peripheral nervous system

Everything else is fluff as far as consciousness is concerned

xris;79651 wrote:

Lets see if they can make a worm and then make it desire a lettuce leaf before they make such outrageous claims.I look at what they have done and then judge them on what they think they can.

Well, you're making an appeal to ridicule and you've also revealed a profound ignorance of the science involved. I would definitely reevaluate this judgment

1 Reply

@odenskrigare,

I see no problem with scientists doing their thing in their lifetimes whatever they wish to do. The issues and concerns that I have is when people try to enforce their fervor on others, particularly through legal apparatus - e.g. the lobotomies of old. I found this in a book written about Jung by Robin Robertson:

"The experience of the "god within" can take many different forms, from a literal belief that one has encountered Jesus or Buddha, to the equally religious fervor characteristic of social reformers , or even scientists. Jung often pointed out that the zeal of materialist scientists to discover the ultimate secrets of the universe is essentially an unacknowledged religious belief."

Now we have a situation where a computer meets the human consciousness. The difference between the two is of a scale that can only be measured by the richness of the universe itself. What should one do when one encounters the "terrifying energy of the Self" [Robertson]. Jung suggests: kneel, bow your head, but don't quite let your forehead touch the ground.

I take that to mean to acknowledge the overwhelming nature of the universe, but also to acknowledge the greatness in oneself. Heroic humility.

Rich

1 Reply

@richrf,

richrf;79653 wrote:

"The experience of the "god within" can take many different forms, from a literal belief that one has encountered Jesus or Buddha, to the equally religious fervor characteristic of social reformers , or even scientists. Jung often pointed out that the zeal of materialist scientists to discover the ultimate secrets of the universe is essentially an unacknowledged religious belief."

That's just his say-so and, like any other psychoanalytical bung, it has no visible means of support

See, it's easy: "The zeal of psychoanalysts to put down inquiry stems from an unacknowledged crack cocaine habit."

True? Hell no. That is only sometimes true. What I wrote there was just a bald assertion. Don't you see how unproductive making things up out of whole cloth is?

richrf;79653 wrote:

Now we have a situation where a computer meets the human consciousness. The difference between the two is of a scale that can only be measured by the richness of the universe itself. What should one do when one encounters the "terrifying energy of the Self" [Robertson]. Jung suggests: kneel, bow your head, but don't quite let your forehead touch the ground.

I don't even know what this means

0 Replies

@odenskrigare,

I'd rather not get caught up in this debate, but I shall respond to odenskrigare....

odenskrigare;79217 wrote:

Reverse engineering - Wikipedia, the free encyclopedia

Reverse engineering implies the ability to *understand* the mechanism of whatever one is reverse engineering, or otherwise have prior knowledge of the principles, and thus come to understand how those principles are manifest in the technology at hand. Common examples would be computer software, or car technology perhaps. But what usually happens here is that researchers are simply "filling in the gaps" between knowledge they already have, and the technological implementation that another team made... the conceptual differences between the two being quite small.

The brain does not fall into this category - if there is no model for how the brain works at all, since it is a product of evolution not human design, then the task of reverse engineering is much, much harder to do. Not impossible, as I said, just harder than, say, reverse engineering a nuclear weapon or the latest search algorithm.

But, even this, has missed the point. The brain, as a mechanism, can be modelled in such a way as to simplify it conceptually. This simplified model of the brain, where neurons are basically akin to transistors, has been around for decades, but no suitable general model was found. Thus, the Blue Brain project implicitly assumes that there was something missing in the simplified model that can be gained or uncovered by going into as much detail as possible in the modelling (ie. they don't simplify the model, but seek to represent it in as much detail as their supercomputers will allow).

The point I was making was that, reasonably speaking, of course this will work, but what has been gained? Very little. We already have the model of the brain in front of us (scientifically speaking), if we just copy that model into a program, what has been learnt about it? Nothing other than if it works, then the model is correct. Rather, even if they succeed they will *still* have to work out and why the model does what it does... this is "true" scientific knowledge. Why not just cut to the chase and try to work it out without spending millions on copying what we already know?

1 Reply

@Hermes,

Hermes;79670 wrote:

Reverse engineering implies the ability to *understand* the mechanism of whatever one is reverse engineering, or otherwise have prior knowledge of the principles, and thus come to understand how those principles are manifest in the technology at hand.

That's the idea...

Hermes;79670 wrote:

The brain does not fall into this category - if there is no model for how the brain works at all

Except that there are, for example:

Numenta - numenta.com

Hermes;79670 wrote:

Thus, the Blue Brain project implicitly assumes that there was something missing in the simplified model that can be gained or uncovered by going into as much detail as possible in the modelling (ie. they don't simplify the model, but seek to represent it in as much detail as their supercomputers will allow).

The Blue Brain project doesn't have much to do with building better ANNs, although such may be useful byproducts of the same. I have no idea what you're talking about now.

Hermes;79670 wrote:

The point I was making was that, reasonably speaking, of course this will work, but what has been gained? Very little. We already have the model of the brain in front of us (scientifically speaking), if we just copy that model into a program, what has been learnt about it? Nothing other than if it works, then the model is correct. Rather, even if they succeed they will *still* have to work out and why the model does what it does... this is "true" scientific knowledge. Why not just cut to the chase and try to work it out without spending millions on copying what we already know?

Oh ok. So ... can we "log" everything that happens in a human brain now? Including gene expression as it happens? This is news to me.

You wouldn't write posts like this if you were following the news. Please please please find a decent neuroscience RSS feed and follow it.

1 Reply

@odenskrigare,

odenskrigare;79781 wrote:

Except that there are, for example:

Numenta - numenta.com

I am aware of Jeff Hawkins and his work, but equally that there isn't a model described in full there. More to the point, you missed the thrust of what I was saying, which was that if there is no accepted scientifically proven model for the brain (that can be used for synthetic intelligence) then your reverse engineering argument is not applicable.

Rather than reading and quoting people in soundbites, try to put things in context, odenskirage.

Quote:

The Blue Brain project doesn't have much to do with building better ANNs, although such may be useful byproducts of the same. I have no idea what you're talking about now.

Hmm... I was talking about the goals and philosophy behind the BBP... the topic of discussion?

Quote:

Oh ok. So ... can we "log" everything that happens in a human brain now? Including gene expression as it happens? This is news to me.

You wouldn't write posts like this if you were following the news. Please please please find a decent neuroscience RSS feed and follow it.

*SIGH!*
I never said nor implied that gene expression was "logged", in fact I never said anything was logged. Rather than result to patronising ad hominem attacks (which is oh so tempting, I know :letme-at-em:) stick to the arguments presented and address what was written. What do you think of my conclusion that even if the BBP succeeds they will still have to work out why and how the model they created "works"?

1 Reply

@Hermes,

Hermes;79898 wrote:

More to the point, you missed the thrust of what I was saying, which was that if there is no accepted scientifically proven model for the brain (that can be used for synthetic intelligence) then your reverse engineering argument is not applicable.

"Scientifically proven" is a phrase that makes me want to punch babies; it is a total contradiction in terms; it always reminds me of the announcer from those penis pill commercials that say their snake oil is "scientifically proven" to "increase size"; please refrain from using it

Vernon Mountcastle's hypothesis, used by Hawkins, is fairly widely accepted btw. Here's a remarkable example of a result borne out by his particular approach to the brain:

Potential of visual cortex to develop an array of functional units unique to somatosensory cortex -- Schlaggar and O'Leary 252 (5012): 1556 -- Science

Visual and somatosensory cortices exchanged in neonatal mice retrain themselves. His model's predictive power is pretty awesome.

Anyway, what kind of model do you want anyway? A model of what?

Hermes;79898 wrote:

What do you think of my conclusion that even if the BBP succeeds they will still have to work out why and how the model they created "works"?

That's too nebulous for me to address

Indeed your last post was also too nebulous which is why I couldn't really respond to it too well

If you're talking about stuff like the role of glial cells in memory, then presumably the Blue Brain project, which, unlike a human brain, will afford researchers the opportunity to observe and poke everything that happens, as it happens, will help do a lot towards resolving those issues.

It would be comparable to being able to observe and manipulate in minute detail anything and everything that happens in a living human brain which we obviously can't do now.

In the particular example I mentioned, it's not as though we can just go into a person's brain, or even a rat's brain, and kill off glial cells that seem to play a role in memory, then restore them, then disable certain functionalities and leave others on, etc. like we can in software.

Even primitive connectionist neural models have already greatly contributed to neuroscience; if you knew this you wouldn't have raised these objections

1 Reply

@odenskrigare,

OK, to make my complain more explicit, I'll give an analogy.

Imagine we take a simple car back to acient Greece, give the smart guys some tools and ask them to find out how it works. Now they can run the car, open up the hood and see that moving parts in the engine translates to movement of the wheels. They might disconnect some parts and thus work out what they do. They might take apart the engine block and find the pistons and the valves and work out that a force in the cylinder drives the pistons and the drive shaft and so on. If we then gave them a small metalworks and more tools, they might be able to copy the car we gave them. What understanding did they gain? Nothing.

To observe a thing, to see what it does is certainly useful, but this is not understanding of what the thing is, observation in itself is not productive. The Greeks need to understand physical concepts like Hoyle's law and carbon chemistry to understand the principle of the engine... *then* they see how to make their own, how to make rotary engines, engines with more or less cylinders, and so on.

This is directly applicable to the situation we are in with the mind. You can talk about NCCs, cortical organisation, glial cells and a great many other (fine) observations of the brain till the cows come home, but these are utterly useless without understanding the principles that underlie what they produce when working as a whole. The BBP is just like the Greeks copying the car, they don't understand how it works, fundamentally, they just observed some functions and reproduced them as best they could.

1 Reply

@Hermes,

Hermes;79919 wrote:

To observe a thing, to see what it does is certainly useful, but this is not understanding of what the thing is

Good point, what is the brain? What does it do?

Could the Greeks, who built the Antikythera mechanism, really figure out what a car is used for by observation?

Could modern day researchers, who had no idea what that corroded piece of s**t was used for when they dug it out of the sea, ever really figure out what its purpose was?

Inquiring minds want to know

Hermes;79919 wrote:

This is directly applicable to the situation we are in with the mind. You can talk about NCCs, cortical organisation, glial cells and a great many other (fine) observations of the brain till the cows come home, but these are utterly useless without understanding the principles that underlie what they produce when working as a whole.

We have known enough of the principles to implement them in commercial products since the eighties, and are learning more and more about them every year, to the extent that genuinely biologically-inspired ANNs are swiftly becoming a reality, e.g.

This is your grid on brains - Missouri S&T News and Events

To say that modern neuroscience can't see the forest for the trees is a reckless exaggeration. Indeed, more theory is needed, but ... like ... to say that we don't know anything general about how the brain works ... that's ridiculous. And one of the aims of Blue Brain is to fill in some of the theory gap:

Frequently Asked Questions[INDENT]The past 50 years have yielded an immense amount of information about the brain, the neurons it contains, the molecules that make up the neurons, and the genes that produce the molecules. There is still a tremendous amount to find out, but we now need a platform where all this information can be integrated and that is the main purpose of building the NCC. "Everything" is a lot, but we have definitely entered a phase of brain research where the brain's secrets are being revealed at an extremely rapid pace. I call this phase the Synthesis Phase where the fragments of knowledge are being collected and assembled to reconstruct the manner in which the brain works - the Blue Brain Project is just one part of this process.[/INDENT]All the information we have now is starting to come together like drops of mercury.

2 Replies

@odenskrigare,

My argument was and still is,it will be more than ten years before even a portion of the brain,mammilian or even more ambiticous the reptillian, can be replicated.It then remains contentious if it can ever be said to be trually a human brain.It would have to contend with the fact of not being human and with no historic memory.This is where my beliefs tell me it wont be conscious because our "I" location has not been found,it will just be another example of a complex computer

1 Reply

@xris,

xris;79969 wrote:

My argument was and still is,it will be more than ten years before even a portion of the brain,mammilian or even more ambiticous the reptillian, can be replicated

Why

btw, you skipped the midbrain entirely and apparently didn't know that the neocortex is the largest and most complex part of the mammalian brain

1 Reply

@odenskrigare,

odenskrigare;80009 wrote:

Why

btw, you skipped the midbrain entirely and apparently didn't know that the neocortex is the largest and most complex part of the mammalian brain

You forced me to read the brief history of the brain and your claims of a human brain being created in ten years is a gross exaggeration.The reptilian brain holds the basic instincts and your mammalian holds the extras.It appears to be the grinning bit but not the aggressive portion, it wont appear anything like me or you.

1 Reply

@xris,

xris;80014 wrote:

You forced me to read the brief history of the brain and your claims of a human brain being created in ten years is a gross exaggeration.

Why

richrf;80016 wrote:

The reptilian brain holds the basic instincts and your mammalian holds the extras.It appears to be the grinning bit but not the aggressive portion, it wont appear anything like me or you.

Ok, but Blue Brain intends to replicate the entire thing

1 Reply

@odenskrigare,

odenskrigare;80020 wrote:

Why

Ok, but Blue Brain intends to replicate the entire thing

If thats the case i have read a different report to yours.Your telling me it has the purpose of all the facilities the body and mind requires to function as a human without the actual body?

2 Replies

Artificial brain 'ten years away'

Related Topics

Quick Links

My Account

able2know