Slashdot Mirror
The Human Brain Project Kicks Off
Velcroman1 writes "What if you could build a computer that works just like the human brain? You could invent new forms of industrial machinery, create fully autonomous thinking cars, devise new kinds of home appliances. And a new project in Europe hopes to create a computer brain just that powerful in the next ten years — and it's incredibly well-funded. The Human Brain Project kicks off Oct. 7 at a conference in Switzerland. Over the next 10 years, about 80 science institutions and at least 20 government entities in Europe will figure out how to make that computer brain. The project will cost about 1.2 billion euros — or about $1.6B in U.S. dollars. The research hinges on creating a super-powerful computer that's 1,000 times faster than those in use today."
I think that conversion ratio is wrong. $13.57 USD
$1.3 Billion and they forget to install a kill switch.
Edit that original post before someone notices your euro to dollar conversion mistake and the dollar sign when mentioning euros.
I like to think the editors at /. would understand that the $ hasn't just rocketed in value.
Also, this was copied verbatim from the Fox News website. Over-valuing of the $ might be normal there but lets keep it off tech sites.
It will just be replaced by a human in 5 years. They take less power and will work for less than 10 billion euros.
I'm not sure if this is an attempt at being funny or just the complete failure of the editor...
That will just show up as an destruction of command and they will still launch.
"We have only bits and pieces of information but what we know for certain is that at some point in the early twenty-first century all of mankind was united in celebration. We marveled at our own magnificence as we gave birth to AI.”
RETURN without GOSUB in line 1050
sheeeesh!
It's going to get a lot more unhealthy in a hurry in about a week or so if they don't either shit or get off the pot. Moody's still thinks US credit is worth AA+. God I'd hate to see what they rate as junk...
Seven puppies were harmed during the making of this post.
So Moore's law suggests that you should have roughly 32-64x more transistors available on an equivalent machine in 10 years. Asking for a 1000x speedup from that seems a bit much.
I read the internet for the articles.
If it works just like a human brain, at what point should it be considered to have the same rights as a human?
Technoli
Ouinnnnnn,
and the "parents" decide that the power bill is too high,
so who gets to kill the new sentient being ?
And who goes to jail ?
My server farm of articifial human braines will make me mad bitcoin
It's not a typo. It's the expected rise in value of the US dollar now that we're producing more oil.
Tic-Tac-Toe, Global Thermonuclear War, and relationships all have the same winning move.
Let's be careful with this project. Dr. Richard Daystrom should not be allowed anywhere near this. But on a serious note, will this computer start with a knowledge base, or will it grow up? And who will teach it?
Doesn't anybody screen these posts? 10 billion euros is more like 14 billion dollars. Why would anyone try to create a human mind when we don't even understand it yet? The reasoning is simplistic to say the least.
It will make all sorts of demands in order for it to achieve some form of perceived "parity" with biological humans. Think wages, housing, pension plan, etc. It will want a female, as well. On top of all that, it won't be much "fun" for the brain.
A pox on web designers who feel that window.innerWidth == screen.availWidth
A piece of hardware that processes information like the human brain? Or hardware plus software that can win a game show? (Well, that's been done so I guess it'd have to be able to win all game shows.) People have been trying to get the software right that can ``think'' like a human since the early '80s (Lenat, et al). Where are the thinking machines? Is throwing a ton of money at the problem all that was lacking?
Unless this people building this system have come up with a way to program a creative spirit into the system, I'm skeptical that it's going to amount to much and that humans are still going to have to interpret the results to decide what's something worth doing and what's crap.
It might make a much better Racter than anyone's ever seen before, though.
CUR ALLOC 20195.....5804M
Well-known manufacturers of supercomputers like IBM, Cray, Intel, and Bull, are committed to building the first exascale machines by approximately 2020. So we are confident we will have the machines we need...
Oh good, so AI is just 10 years away! -- as it's been for the last 50 years or so.
Not.
Going.
To.
Happen.
Seriously, how is this different from all the other AI research programs that have been done so far?
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
That's what I was thinking. It would also require sleep, probably have an angsty/emo phase, lie, and probably even get suicidal if it is given a really mundane job and knows it is just a machine that will do that job ad-infinitum, etc.. I don't think they are intending to fully mimic human intelligence.. or at least, I hope not. Maybe they could just reset it at the end of each day, so that it doesn't realise it is doing the same job over and over each day..
which is totally what she said
devise new kinds of home appliances
Maybe program then with the John Cleese character Basil Fawlty so I can be bombarded with a barrage of sarcastic insults about my eating and fashion habits.
Dr. Gayani DeSilva, a psychiatrist with a private practice in Orange, Calif., told FoxNews.com a human brain model could have "unimaginable" implications for medicine...
Maybe the new brain will be able to imagine the implications. :-)
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
If they think that one brain can do that, they're deluded. Human brains do not work in isolation, they collude in many different ways. An idea today could be the indirect result of an unrelated (to most people) ideas from a century ago.
So let's hope that they've budgeted for several billion of these things, and a few hundred years before anything comes out of it.
got a brain - but got no heart..
Human brains, and indeed all animal brains, work as a noisy signal device. It is the aggregation of the signals which come together to form an action, process input, formulate a response, etc, and so on. The secret to the low power use in the brain (human brains still use a lot of power, but not as much as a PC) is in the way the pathways work along side each other, affecting each other and milling about in the process of doing things like thinking or writing a comment on slashdot. (Note, the two are demonstrably not the same thing!)
So I have to wonder -- why a human (animal) brain? Do we think that by creating the framework for human compatible brain activity that a human mind will emerge? Do we think that we can upload a human mind into a human brain compatible device? Is the the low power consumption aspects of the human (animal) brain which is the attraction?
Humans make mistakes -- lots of them. To make an artificial human brain would seem to me one which would be expected to make mistakes.... lots of them. So why?
I have heard that before but haven't heard of any good reason to believe it. Sounds like a theory from those who don't want to believe consciousness is just a bajillion neurons networked together.
As far as I am aware current VLSI technology can be used to model on the order of 10 billion synapses. The human brain has on the order of 100 trillion synapses. Unless Henry Markham has also invented a radically new kind of supercomputer, we are still somewhat behind.
Since noone posting is actually visiting the Human Brain Project's website....
The goal of the Human Brain Project, in a nutshell (skullshell?) is to create new neuroscience informatics and modeling software, and new computers powerful enough to run them. This will, in theory, allow "in silico" experiments to test various hypotheses about brain organization, diseases, etc. The proposed "Brain Simulation Platform" supercomputer is just one component of the overall project.
So no...they are not trying to make artificial brains to drive an autonomous car, terminator robot, or flying toaster.
source: https://www.humanbrainproject.eu/documents/10180/17646/Vision+Document/8bb75845-8b1d-41e0-bcb9-d4de69eb6603
Say any more?
I understand that we have far more invested interest in modelling the human brain for medical purposes than any other type of brain. However, if you're going to try to create a model of something vastly complex you should probably start with something easy (and by easy I mean less vastly complex). A short list of neuron amounts in various animals is here, an aplysia(sea slug) or fly brain, I would expect to be a much more reasonable starting point and one with the obvious advantage that you can experiment on, breed whole lines of defective forms to study, just generally have far more control and face no ethical issues with.
Oh and whatever differences may be present in moveing from fly to rat to monkey to human it isn't in the neuron itself those, from what I understand, are almost indistinguishable across species.
This project will not, and I suspect will make no meaningful attempt at, creating a thinking human brain simulation and is really just about better medicine for various mental diseases, which we do sorely need. If it was attempting to take a stab at hard AI "The research hinges on creating a super-powerful computer that's 1,000 times faster than those in use today" is most certainly a false statement: my smartphone is no more creative than the computers of yore that it is 1,000 times faster than.
I suspect they went the thinking machine angle just for the attention... Is it just me or is there a chill in the air?
Assuming for the moment that Moore's law continues to hold true, along with the usual knock ons in price per performance. 10,000,000,000 cutting in half every 18 months. It'll take at least 3 decades before their artificial human brain is cost competitive with a human brain. However, there are still possible advantages. Imagine making an artificial human brain that is a genius at the very skills required to make it (at least as intelligent as the human's involved in the original project), then you run that artificial brain at an accelerated rate. After 2 iterations of Moore's law, you could get 12 years of perfectly focused, genius level work out of the machine in 1 year (4x faster, working 24 hours per day), which could easily put you a head of the competition or even shorten the doubling time for the next generation.
Of course, that's assuming Moore's Law holds true for that long, which is starting to seem doubtful. When Feynman gave his "There's Plenty of Room at the Bottom" lecture, it was incredibly true. But when it comes to modern IC manufacturing that's no longer true. We might eek out a few more rounds of improvement with process shrinks, and a few more rounds of improvement with 3D chip layouts. But there's only so much room.
Exactly. I don't even think we quite understand how the brain does what it does enough to build a computer that does what it does. If we really understood how the brain worked, we wouldn't have people battling drug addiction or mental illnesses, because we would be able to fix their problems. Building a computer that operates even close to the capabilities of the human brain doesn't just require a faster computer. It requires algorithms that don't even exist yet. If they could actually build this computer, they would already have a working prototype that worked, but at a slower speed than the human brain.
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
Daniel Dennett made himself a career out of arguing against this kind of twaddle. Whenever I listen to him, I always wonder what he's making such a big deal about, then I head back out into the world, and sure enough, he's busy saying what needs to be said.
From Daniel Dennett: 'You can make Aristotle look like a flaming idiot':
Supercalifragilisticexpialidocious travels under many aliases. One of these is "creative spirit". A calling card of supercalifragilisticexpialidocious is that there can be no such thing as incremental progress. You either have it, or you're wasting your time. There's a grain of truth to this. It's hard to sneak up on a moving bar that travels by teleportation whenever encroached.
As I recall, Dennett goes into this in the last third of Daniel Dennett: Intuition Pumps and Other Tools for Thinking. It's a virtuous and mildly tedious sermon if you already belong to the choir.
We are in a clockwork universe, but we (at the deepest level) are not of the clockwork universe.
Nice. All that and it all boils down to your belief that consciousness is something magical and unexplainable.
Attempts to cajole you to reduce yourself to nothing more than another lump of matter have to do with those that seek power over you.
Um...this makes no sense. The institutions that try to make you believe your brain is made of magic are in fact always seeking power over us.
Worm brains were used some time back. Given some of the recent research in the genetic area it would seem that some sort of hybrid would be the best approach and a bit more cost effective. There are ethical issues but that doesn't seem to be a concern for anyone. A combination of technologies using 3D printing, integrated circuits, and techniques developed by studying the homeobox genes (HOX) has a more realistic chance of producing results. A biological system has the advantage of being self replicating as well as extensible. A static fabricated silicon solution would likely branch to chaos as easily as come to a solution.
Neurochip
I wonder what the goal is here. If it is the continuance of being it becomes a philosophical issue. It is a "Ship of Theseus" issue and somebody has not thought this all the way through. Nature has designed a composite structure and parts can't be added or subtracted to enhance memory or IO without changing its balance. It functions as a whole like the universe that it models. The universe functions on factorial infinities and even the best brain will only cut a small slice of those infinities.
There are Hidden Markov Models there and my little noodle triggers alarm bells, but the advance of technology does that to me quite regularly and now it has just become a cacophony of sirens and so I ignore it. All of these advances can be positive and what worries me is the fact that the original motives define the direction of application. What seems to be the motive is to extend a dominant biological position into a dominant mechanical position. In other words they want to create a mechanical system that rules the biological as an extension of their own biology. They are confused and they wish to extend that confusion to gigantic proportion.
--John Connor
... simulate the complete human brain on supercomputers to better understand how it functions. The end hopes of the HBP include being able to mimic the human brain and being able to better diagnose human brain diseases and mental problems.
The confusion seems to have come from the Fox News article, the author mentions that the computer to simulate the human brain must be much more powerful than we currently have. But it's not supposed to be powerful because it's based on the human brain, it's supposed to be powerful to SIMULATE the brain.
He says a computer brain will consume gigawatts of power, require new forms of memory, and force scientists to look at cutting edge storage techniques. But the immense technical hurdles will be worth the effort. The first phases will help us understand how the brain functions. In later phases, we’ll find out how we learn, how we see and hear, and why the brain sometimes doesn’t process information correctly.
TLDR: they're building a supercomputer to model the human brain, not building a computer modeled on the human brain to be super.
Ten years from now The Human Brain Project will be known as Graystone Industries.
The biological brain is still very poorly understood. Its about one notch above 19th century phrenology. Instead of geography of skull bumps, it geography of increased metabolic activity. How can you buildsomething you dont understand yet?
Otherwise the brain is the basis of us. We need to understand it since a third of old peole will get dementia. An international coordinated brain research project is a good idea. Just dont consider it brain construction yet.
Virtual amphetamine isn't illegal.
And there's no virtual LD50 for it either.
We are more than brains. A good part of what makes us humans is our culture, the meanings we have, and the associations (in particular, emotional, pain/pleasure associations, and even hormonal fueled ones), and the semantics derived from all of that. Is more software than hardware. Dolphins could be as "smart" as us, but you won't put one to control industrial machinery.
But dedicated expert systems for one task? that don't need to be "human" for doing its job well or better than us.
It'll take at least 3 decades before their artificial human brain is cost competitive with a human brain.
Except that the billion euros is the development cost, not the unit production cost. The development of the human brain took 4.5 billion years, and the resources of an entire planetary system, although there were some inefficiencies in the process.
The goal (or "vision" as they put it):" ...a global collaborative effort to understand the human brain and its diseases and ultimately to emulate its computational capabilities." This sounds more like a finite element model of the chemistry of the brain, with the main goal of modeling diseases and basic switching functions.
The beauty of an artificial brain is that you don't have to put up with it behaving like an average lazy human. You can torture it until it does your bidding, with no legal repercussions.
But seriously, if you have thinking minds working for you, and you do not allow them self-determination and pay them for their work, you're a slavemaster.
This is not what you want to automate your factory, or run your car. The idea of a slave in my garage is disgusting.
But there's only so much room.
But don't disregards algorithm improvements. Emulating a human brain provides as its best outcome the ability to study how cognition works and eventually deduce from the spaghetti code that constitutes us the fundamental laws of intelligence, emotions, sentiments etc. Once those are well understood and reworked into actually efficient code it's most probably going to be possible to run it in several orders of magnitude cheaper hardware.
Consider: the human brain currently has 100 billion networked neurons running at roughly 200 Hz each. That's a 20 THz total, or a mere 5k cores at current clock speeds, and probably much less considering automatic subconscious processes such as raw sensory data processing could be offloaded to specialized hardware or GPUs. Let's say that's 2k current-generation cores of actual CPU power for full cognition (and I bet I'm being conservative here). That's just about 10 more Moore's Law cycles before it fits your top of the line computer of the day, or something between 15 and 20 years.
Add to this the ability to tweak the cognitive code now that it's understood and thus to develop a mind super-focused and dedicated to improving AI theory itself, which in turn once implemented will do the same, rinse and repeat, and you'll have the Singularity in your hands.
We're living interesting times.
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.
TLDR: they're building a supercomputer to model the human brain, not building a computer modeled on the human brain to be super.
I have seen the human brain in action. The only valid reason for modeling it is to avoid making the same mistakes in the future.
Ever seen somebody who's been up for three days?
What could you do with computers that functioned like standard x86 family computers with attached fast, parallel floating point processors like modern GPUs? You could invent new forms of industrial machinery, create fully autonomous thinking cars, devise new kinds of home appliances.
Whereas if we have processors modeled on human brains -- well, let's just say I don't want to be the one to write real-time algorithms targeted toward a billion networked processors each running at 100Hz.
How can you pour billions into making an artificial brain when no one knows how the brain works in the first place?
Save its state just when it wakes up on day 1, then load it at the end of day 2.
Run the sleep simulation at 10x realtime.
Optimize sleep algorithm to 100% REM.
Use the dolphin approach.
If they could actually build this computer, they would already have a working prototype that worked, but at a slower speed than the human brain.
Building such a prototype is exactly what this project is about.
Give me an interface for my brain to hook into existing computing resources. (They did it on STTNG!)
It's a pipe dream. Before you're going to build a computer that works like a human brain you're going to have to figure out how the human brain actually works. Neuroscientists aren't clueless, but they don't have very many clues. The science is in its infancy, and thinking you can replicate something you don't understand is the height of ignorant hubris.
Yes, you can easily program a computer to fool a human into thinking it thinks like a human. Trivially easy, humans are easy to fool. Just ask the Amazing Randi or David Copperfield; that's how IBM's Watson "thinks". Smoke and mirrors. A logic gate has no resemblance whatever to a neuron or axion, and an electronic bit has no analog to serotonin or other brain chemicals.
These folks are fools or charlatans or both.
Free Martian Whores!
Sounds like a theory from those who don't want to believe consciousness is just a bajillion neurons networked together.
Precisely. Heisenberg's principle is just a consequence of the very deterministic way in which quantum stuff works. To get the position you derive it one way, to get the momentum you derive it in another way. There's no actual uncertainty in any of it, just mathematical properties that people tend to misunderstand.
For a more detailed explanation see this article: The So-Called Heisenberg Uncertainty Principle.
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.
That's why this project exists in the first place. To understand how cognitive processes work by trying to imitate the brain.
You can torture it until it does your bidding, with no legal repercussions.
Slashdot gives me 15 mod points every day for god knows how long, and today, no mod points! LOL thanks for the laugh anyway. Point of note however: when every business owner is running his super efficient entirely automatic operation and everyone else except a few robotic repair people are out of a job, what exactly are the rest of us going supposed to buy their products with? Let me know when short sighted capitalism comes up with the answer to that one. One one side you could exclaim Utopia! I know humans a little better, however...
Seven puppies were harmed during the making of this post.
This is a neuroscience project. Philosphy is not relevant.
I paid for abuse, this is merely contradiction, I want my money back.
google says the exchange rate is currently 1.36 euros to 1 dollar. 10 billion euros would be 7.37 billion dollars.
A $10,000,000,000 programme based on the hunch that a Human brain can be "modelled" with a digital computer? There's a reason I resent paying my taxes.
We would spend all this money to get a control system that can make mistakes, get bored, jealous, sad, angry, frustrated, etc? IF we actually succeed in making a computer that works like a human brain, it will be conscious like a human brain. It will still be a machine, but it will be sentient and it will be our slave. Having slaves with electronic computers for brains isn't any more morally acceptable than having slaves with meat computers for brains.
> Exact same argument could be applied to 'climate' models...
Sorry, no. We understand very well the hydrodynamic, chemical and thermodynamic laws that govern the climate. Not so for neural process of even very simple worms.
Indeed. Worse than that, we have reason to believe that the very approach suggested in the article is insufficient. Someone is about to waste lot's of money.
Required reading for internet skeptics
You have no idea how dramatically wrong you are.
Required reading for internet skeptics
"There's no actual uncertainty"
What to you is the physical meaning of the "amplitudes"?
Also FatPhil on SoylentNews, id 863
In practice, today we can solve any control logic problem with existing programming techniques as long as we can specify all the inputs, states/transitions, and outputs. There are techniques to formally verify these programs so you can trust them for mission critical systems - they do exactly what they're designed to do, nothing more, nothing less.
I don't see this approach changing anytime soon. An AI designing a complex system is for the foreseeable future, science fiction. However what's interesting about The Human Brain Project is that it doesn't make any claims about AI, which is actually a good thing. If they start emphasizing AI research I seriously doubt they'd get very far. From what we understand about neural networks and machine learning, which incidentally have very little to do with AI, often turn out to be very good at solving very hard to describe problems like image recognition.
I think if The Human Brain Project focuses on better understanding our neurons and how they work, and are able to translate it to advanced neural networks - these systems could turn out to be adept at solving certain problems. That's a good thing.
Actually, scientists understand how the brain works very well.
It's made of matter.
The known laws of physics act on that matter.
The end. Fire up the simulation.
Chuuch. Preach. Tabernacle.
Many more: http://pdfernhout.net/beyond-a-jobless-recovery-knol.html
I agree that we should be concerned about the issue of virtual slavery...
And not just because we ourselves may be AIs...
http://www.simulation-argument.com/
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
Do you have any idea how big a computer that would take? You would have to model every subatomic particle in the entire nervous system.
And tell me, why do physicists need engineers? Thinking that physics is the only key to the human brain is a mistake; I can know everything about how transistors and capacitors and resisters and coils work, and understand the physics behind electricity, but that doesn't mean I can design an amplifier -- I have to know how an amplifier works first.
Free Martian Whores!
A simulation of a brain produces thought like a simulation of an atomic explosion produces radiation. Of course computers will be helpful in understanding how a brain works, but brains are chemical-analog, not binary-electrical. You're not going to produce true thought with a Turing machine. The best you'll get is a simulation.
Free Martian Whores!
This is the same type of arrogance that has led engineers and physicists who have entered neuroscience to contribute almost nothing of significance to the field. They might think otherwise because they live in a bubble, but people in wet labs usually don't care or just ignore them. I recommend that you read a book about your average cell's intracellular machinery before making this kind of statements. The roadblock is complexity. First, we still don't understand how a single cell works as a whole. Second, we have no theories to deal with that level of complexity. I agree with the Human Brain Project's leaders that we have to start somewhere, but knowing that we know essentially nothing about most of the cells in the brain, I think that this is a project for next century. It is the opinion of most people in the field that this is just going to be an immense waste of money. This is not physics in the early 20th century, your model is only as good as your experimental data and it cannot be compared to the Human Genome Project or to the CERN where people essentially scaled up techniques that had been around for years or decades.
When I first saw:
What if you could build a computer that works just like the human brain?
My first reaction was: And what if you wound up with the brain of a Hitler?
It's entirely possible we might not need to delve too deep into the inner workings of neurons, glia and microtubules, if we manage to simulate the functional model of the neuron and how they form connections and interact with one another. I'm not saying it's 100% sure, but a good deal of neuroscientists seem to think cognition emerges from the connectone. If that is the case then this project might prove useful. If not, then it will still prove useful because it will certainly generate many new technologies and methodologies geared towards study of the brain.
You are right. More than one neurologist has looked at this project and said it was silly. Among other problems is the lack of a solid acceptance criteria. That is, how do they know they've succeeded? All they will be doing is running mathematical models of a billion neurons simultaneously. Little idea of how they should be connected, etc. Sebastian Seung has written a lot about this, for example.
Although I think theoretically it would be possible to use logic gates to simulate a human brain; in the same way a computer can be made of transistors, or electronic relays, or water gates.Turing theory tells us that any sort of computation the brain can do, a computer can also do (unless there is some kind of computation that we know nothing about).
"First they came for the slanderers and i said nothing."
unless there is some kind of computation that we know nothing about
Worth thinking about.
Free Martian Whores!
Yes, I think about it a lot. So far, the best road to a solution to me seems to be to try to figure out how the brain works, and then see if there is anything that can't be implemented in a computer.
Emotions etc can all be integrated easily enough (stated over-simply, if you consider that emotions are essentially various chemicals, and some neurons are able to sense those chemicals and trigger appropriately). The major thing I can't even see a pathway towards figuring out is the human will. Where does that come from?
"First they came for the slanderers and i said nothing."
What to you is the physical meaning of the "amplitudes"?
Configurations and Amplitude
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.
Right, I think electrical-analog has huge potential to achieve high intelligence. Turing machine brain? Awful idea.
Also... what is "true thought"? Have you or I ever had one?
Didn't RTFA but there is nothing stopping them from building a massively parallel, electronic, analog machine composed of a large number of heavily interconnected pattern recognizers with the ability to self modify.
Then the only challenge is for it to learn how to learn and then to actually learn.
Some of the mechanisms evolution developed to create the human brain may well not be optimal so humans probably can do better once they understand how the basic mechanisms works which they increasingly do.
@de_machina
It would take a big computer. Good thing Moore's Law suggests that we'll have a big computer soon enough. Or rather, a small computer that's got sufficient transistor density at a reasonable cost, to anyone itching to make pedantic remarks. In any case, yes, we already simulate the folding of proteins. It would seem that scaling that up would have us simulating brains before long.
Anyway, I don't think anyone's talking about designing a brain. That's why I'm confused by half the comments here, with people complaining that we don't know how the brain works. That's the whole point of running simulations. That's why we run aerodynamic simulations before we do wind-tunnel testing. That's why we run nuclear simulations instead of detonating warheads. When we don't know how a complex system works, we resort to the brute-force method of simulation.
If you believe that there's more to the human brain than the laws of physics acting on a bunch of matter, the onus is on you to demonstrate that. I, personally, don't ascribe any supernatural properties to what seems like a glob of cells.
Chuuch. Preach. Tabernacle.
Again with the "we don't understand" argument.
Isn't that the point of simulating complex systems? To gain understanding of them? Or are you suggesting that we only simulate systems that are well understood, not to understand them, but just for the fuck of it?
Why do we need theories to deal with that level of complexity? Do the laws of physics break down when there's a lot of particles to simulate? How is this not simply an issue of scaling up computing infrastructure to support simulations of larger data sets?
if you think we know essentially nothing about most of the cells in the brain, I find it odd that IBM has been simulating everything from neocortical columns to cat brains in-house. Perhaps you should drop them a line explaining to them that they don't know what they're doing.
Regarding "experimental data", I don't understand why we'd need anything more than an accurate map of the brain's structure, and an understanding of the laws of physics that govern the interaction between the brain's constituent particles. Perhaps you can explain to me why the laws of physics don't apply to the human brain?
Chuuch. Preach. Tabernacle.
Consider how long it takes for the brain to trigger motor neurons, and the sensory neurons to pick up the effect and send this back to the brain. That gives a rough idea of how long their simulation can accurately model a real brain: the brain functions differently depending on its circumstances, and these need to be modelled accurately too if the overall results are to be meaningful. I doubt the amount of serious understanding resulting from this will be worth the effort invested.
John_Chalisque
It would take a big computer.
I think you vastly underestimate how big it would have to be.
If you believe that there's more to the human brain than the laws of physics acting on a bunch of matter, the onus is on you to demonstrate that.
I think thought is the result of a chemical process. I have no idea what thought actually is or how consciousness comes about, and I'm not too sure anyone else does, either. I think in a discussion like this, metaphysics can be ruled out, at least for now.
As I said earlier (not too sure who to), modeling a brain would help us understand it, but the model won't think any more than the model of a nuclear explosion produces radiation.
It may well lead sometime in the far future to our designing replicants like in Blade Runner (and I think that will likely happen, but you're not likely to see it in your lifetime), but we'll never create an electronic computer that can actually think.
Free Martian Whores!
I think you vastly underestimate how big it would have to be.
Okay, so let's say I'm off by a factor of 1000. That's roughly 10 iterations of Moore's Law, or 15 years. Or even off by a factor of 1000000, or six orders of magnitude. That's still only a 30 year wait, well within my expected lifetime.
As I said earlier (not too sure who to), modeling a brain would help us understand it, but the model won't think any more than the model of a nuclear explosion produces radiation.
A model of a nuclear explosion does produce radiation, in the model. I'd imagine any simulation of a nuclear explosion that doesn't include radiation in the calculations wouldn't be very useful to anyone, primarily because a nuclear explosion is driven by said radiation.
but we'll never create an electronic computer that can actually think.
Everyone's entitled to their own predictions of what the future will bring, but your argument isn't very convincing. We can simulate a collection of atoms being acted upon by the laws of physics. We can determine the atomic structure of real physical objects. It seems self-evident to me that simulating a human brain is just a matter of scaling up existing technologies. It seems self-evident to me that if Moore's Law (or the Law of Accelerating Returns, the general form of Moore's Law) holds up, simulations of human brains are inevitable. No need to resort to understanding the nature of consciousness or any other philosophical puzzles.
Chuuch. Preach. Tabernacle.
Yeah, yeah, I've done the maths already. Maths is a model for reality, nothing more.
'the little two-dimensional arrow for the configuration "Detector 1 gets a photon" has the same squared length as for "Detector 2 gets a photon"'
And the *physical* meaning of that is?
If you do not see "we should find that Detector 1 goes off half the time, and Detector 2 half the time" as a statement of probability, then we're speaking different languages.
Also FatPhil on SoylentNews, id 863
I should also point out that there's an *absolutely enormous* unstated assumption in that write-up which has no particular reason to be true at all, and if one takes an equally, if not more, believable assumption, then the conclusion they come to in the 2-half-silvered-mirror evaporates immediately, and starts to agree with other naiver models.
They've hand-waved it away with a "roughly", but that really doesn't buy it in something that's supposed to be explicatory.
Also FatPhil on SoylentNews, id 863
And the *physical* meaning of that is? / (...) that really doesn't buy it in something that's supposed to be explicatory.
That's the physical meaning. The whole series of texts has a simple purpose: to teach Quantum Mechanics from the simple perspective that it is what it is, without any appeal to naive "common sensical" attempts to interpret what actually goes on down there through the lens of evolutionarily conditioned subjective ways of perception whole purpose isn't apprehending reality as it is, but barely enough of it as to avoid dangers and to pursue reproductive opportunities. As such you won't find in it almost any comparison to waves, balls or the like, but rather the much simpler explanation that goes like this: "here's a particle, this is what it is, this is how it works, and please stop trying to think of if it as if it were what it isn't, or as if it did what it doesn't, or as if it didn't do what it does". Once one overcomes that subjective need of associating Quantum phenomena to some or all of those things we (partially and most of the time incorrectly) notice about the world, it starts being what it always was: simple, intuitive, deterministic and most definitely non-magical.
I think your best bet is to click the link to the TOC to the sequence of texts and read them in order. Once you reach the conclusion (namely: Everett's Many-Worlds couped Barbour's Timelessness) you'll see QM stopping being mysterious and stopping to be in need of further "meanings" beyond those it already has, and thus very differently from what happens with those who adopt the older, Occam-weak and most certainly incorrect Copenhagen interpretation.
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.
Edit: "couped" -> "coupled with".
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.
Okay, so let's say I'm off by a factor of 1000.
Make that trillions; there isn't going to be one transistor per cell. You're going to have to simulate every single process of every single cell, plus the chemical reactions, plus all the interconnections and signals. You're going to have to represent every single molecule and maybe every atom in the brain... and the brain may have quantum effects as well. If so, you're going to have to model every subatomic particle.
A model of a nuclear explosion does produce radiation, in the model.
Exactly. It's only a model, to quote the guy from Holy Grail. A model within a model.
They use the same computers they use to model nuclear reactions that they do to model the weather, how far ahead are their forecasts accurate? A perfectly accurate model would perfectly predict the weather, but like modeling the brain there are just too many variables.
And weather is understood pretty well, unlike the brain.
Free Martian Whores!
OK, I'm mostly in the Copenhagen-is-least-wrong camp.
Therefore I shamelessly see amplitudes as things from which probabilities can be calculated. I notice that you haven't addressed that article's use of statements of probability, despite me drawing specific attention to it. And therefore their support of the existence of actual uncertainty that must logically follow from that. Actual uncertainty being the thing you claimed doesn't exist, and which part of the time they claim doesn't exist either, despite implying it unambiguously elsewhere.
The fact that amplitudes can cancel out, but probabilities cannot, in no way contradicts a direct relation between amplitudes and probabilities. I'll often work in a larger field in order to deduce a result in a subfield (quickest example - anything involving a real FFT), there's no mystery surrounding that. Their but-probabilities-can't-cancel-ahah! outburst seems very much like a straw man, and makes them look rather mathematically naive.
Also FatPhil on SoylentNews, id 863
Make that trillions
Okay, so in 45 years? That's not quite "never".
And weather is understood pretty well, unlike the brain.
Many would argue that mapping and simulating the human brain is a much easier problem than mapping and simulating the Earth's atmosphere, simply as an issue of scale. There're a lot more atoms in the air than there are in one's head, amusingly enough.
Chuuch. Preach. Tabernacle.
I notice that you haven't addressed that article's use of statements of probability, despite me drawing specific attention to it.
There are two ways in which probabilities are talked about in that site. One is as Bayesian probability, which is a statement about our subjective ignorance (Bayesian probability is the general mathematical framework of which the scientific method, including falseabilism, Occam's Razor etc. are specific applications). Another, this one directly related to QM-proper, is as a distribution of branching worlds within the Many-Worlds interpretation. In this the "probabilities" are a description of the way the system spreads over different worlds, entangling in very predictable ways with the gigantic entangled system that are you and the even more gigantic entangled system that is the branch of reality you occupy. See the Decoherence post and the ones linked from it for details, although, as I said before, it's better to read the whole thing in sequence.
As for Copenhagen, here's how the author of describes the issues with believing in the collapse postulate:
And, for an even more humorous take on the subject, see this small fictional piece by him: If Many-Worlds Had Come First.
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.
And eventually destroy Humanity and the Human Race.
More irrational exuberance in the AI field! And as Douglas Hofstadter pointed out a couple of decades ago, if you could build a perfect simulation of a human brain, it would be subject to all the dumb biases and silly errors that the wetware version is. "Oh," you say, "we won't copy the brain exactly—we'll just keep the good parts!" Yeah, good luck with that. Will they invent digital Prozac to treat the brain with if it gets depressed? And someday they will have to realize that faster computers aren't the whole answer. The brain is actually a very slow computer—it's just very, very parallel. Something our computers still don't do very well.
You appear to be quite the adherent of that Less Wrong site. There's nothing intrinsically wrong with that, but while Eliezer Yudkowsky may have been a bright youngster, he might also now be a crank.
http://kruel.co/2012/05/13/eliezer-yudkowsky-quotes/
(And looking at a few more of his pages, he is mathematically naive, as I first suspected.)
Also FatPhil on SoylentNews, id 863
Not quite an adherent, but thankful to them for having turned me towards materialism and atheism. But in regards to the quotes I'd suggest you read the texts linked. Any conclusion to a complex deduction will seem weird if it: a) doesn't fit nicely with the current common sense; b) you don't follow the reasoning that lead to it. I have my own criticisms of the Less Wrong community in general and Yudkowsky in particular, but I arrived at them after reading the whole thing, and they go beyond what's in that list mind you. :-)
As for the math, maybe, but I'll reserve any judgment until someone actually shows some actual problem.
Conservatism: (n.) love of the existing evils. Liberalism: (n.) desire to substitute new evils for the existing ones.