What is Digg?98 Comments
- oesj, on 10/12/2007, -0/+83and the average youtube user needs a lemon and some copper wire!
- TexanPsycho, on 10/12/2007, -3/+61@Ngai: The average Digg user needs a basic 800MHz processor to emulate their brain, with at least 128MB of RAM and a 4x CD-ROM drive.
- Ngai, on 10/12/2007, -1/+58Give it a year or two...
we'll get the other half. Thats pretty amazing if you ask me...
Imagine...
If they tried to emulate your brain....how many "super computers" would you need? - number5, on 10/12/2007, -6/+61a super computer and emulate Only half a mouse brain
- FizixMan, on 10/12/2007, -10/+44The problem lies in that our sense of "computing" is vastly different from how a brain works. Whereas brains naturally work as... well work as brains, we have to try to emulate those processes with our given ability to program. Which unfortunately, is programmed on systems that don't work like brains.
No matter how powerful a super computer is, it really comes down a series of logical instructions. True, false, or, and... etc.
Which, when you think about it, is pretty dumb. Imagine describing life that way. - Zzyw, on 10/12/2007, -9/+37Fizixman: Neurons fire or they don't. 1 or 0.
- FLaw, on 10/12/2007, -2/+22Ghost in the Shell only 30 years away?
- zeromancer, on 10/12/2007, -0/+19"Rajagopal Ananthanarayanan"
.... seriously ? - lndmn01, on 10/12/2007, -0/+13FTA : "The researchers say that although the simulation shared some similarities with a mouse's mental make-up in terms of nerves and connections it lacked the structures seen in real mice brains."
So they say that they ran a simulation of half a mouse brain but in reality they ran a partial simulation, of a simulation, of half a mouse brain...
WTF did they do? - thirdoffive, on 10/12/2007, -1/+14I for one welcome our tetrapedal-cybernetic-cheese eating over lords.
- Eric4, on 10/12/2007, -1/+13"The vast complexity of the simulation meant that it was only run for ten seconds at a speed ten times slower than real life - the equivalent of one second in a real mouse brain."
- GiggleStick, on 10/12/2007, -3/+15zzyx:
That's not true. Neuron firing is not simply on or off. There is a rate of firing that makes a difference, and probably the strength of the firing matters too. Also the way they are "networked" makes a difference. There can be multiple neurons feeding into a single dendrite of another neuron, etc. It's a little more analog than you might think. - notjamt9000, on 10/12/2007, -1/+12Would go well with this: http://www.instructables.com/id/EM1O9XJF0YPF7M5/?ALLSTEPS
But seriously, this is an amazing achievement. Imagine converting real human brains to ROMs and emulating them, storing all the great minds of our time. Now they need to simulate a cat brain and see how it gets on with the mouse. - edzieba, on 10/12/2007, -1/+12Less than that. Direct neural interfaces are already in clinical trials. The GiTS cyberbrains were not brains simulated on computers, they were ordinary brains with augmentation and direct links to external devices.
Unless you mean Project 2501, which wasn't a simulation of a brain but something else entirely (a 'ghost in the machine'). - TrainingName, on 10/12/2007, -1/+12STOP MAKING AI
WE'RE GONNA BE OWNED BY TERMINATOR ***** MICE - bIuebonics, on 10/12/2007, -1/+10arg... discrete*
- Logikos, on 10/12/2007, -1/+10Smartest... or richest. If you were to create a form of afterlife for people where they can run around in a 3D universe and interact with their friends, family and distant descendants for generations... well, imagine the possibilities. I know quite a few people that would willingly die prematurely to be digitally reincarnated into World of Warcraft.
- inactive, on 10/12/2007, -0/+9its not the hardware that im worried about, eventualli science will get there to match the human brain
the software on the other hand is gona take a lot more genious thinking to figure out - donjaime, on 10/12/2007, -0/+8Basically, their simulation isn't totally true to the actual physical reality. There are almost certainly processes going on in the real world that they have not accounted for, or have incorrectly modelled. They need to keep working on the model for a few more years to get it closer to the real thing. Once they get the model driving the simulation "right" it should start looking more and more like what they see in actual living mouse brains.
- hendrixelixir, on 10/12/2007, -1/+9@gigglestick:
The rate of firing may make a difference, as well as which neurotransmitters are used to send the message along the synapse between two neurons, but the strength of the signal (polarization) is a constant. A non stimulated neuron has a charge of -70 mV, and when the message is sent a large influx of Na+ ions flow in to change the polarity and thus effect the signal. I can see how a supercomputer might take a stab at making thousands upon thousands of connections to make a rudimentary simulation of a mouse brain. - chromie, on 10/12/2007, -1/+9Mr. Samir Naga... Naga... Naga... Not gonna work here anymore!
- Joe_rigby, on 10/12/2007, -0/+6What kind of ethical problems would develop around creating a digital human? Some of us are concerned about abortion and stem cell research today, tomorrow we might be worried about turning off our computers.
- morninglorii, on 10/12/2007, -2/+8Gigglestick beat me to it.
If you'd like to think of neurons in terms of computers, don't think about them in terms of bits (0/1, firing/not firing).
Think about them as linked lists (and each node can be linked to many other nodes), which pass numerical values to one another. The links represent the multiple connections between neurons, and the numerical values represent rate of firing. - FLaw, on 10/12/2007, -1/+7Very true, but does the same go for full cyborgs?
I'm just itching inside talking about a futuristic anime coming true so soon... very possibly in my life time.
Cant wait 8) - inactive, on 10/12/2007, -0/+5If i weren't human, of course i'd want to kill humans. We're a species that kills other species for little or no reason. We destroy the only planet we live on with no thought of our future survival. Destroying all humans would be good for most other species in existence.
- Phoenixfury, on 10/12/2007, -1/+6It's all in good until they accidentally emulate Brain's brain (of Pinky In the Brain). He'd find a way out their private network.. Then he'd find the nearest robotic toy mouse and upload himself into that... Then were all screwed!
//sarcasim - thirdoffive, on 10/12/2007, -0/+5Oh yeah wrong name, well botched that joke, digg down...
- Jugalator, on 10/12/2007, -0/+4It's easy to draw parallels with this experiment and the past pre-transistor ones like ENIAC.
We're basically experiencing tests before a revolution, and struggle for the same reason yet again.
Before, technology was inefficient for mathematical computations.
Now, we fit an ENIAC in a pocket, but a next hurdle is technology being inefficient for neural nets.
This experiment is amazing though, and one thing that's interesting with biotech and silicon is that I'm not convinced biotech will remove the need of today's technology, because they're useful for different needs. To illustrate my point, just consider how hard time a mouse brain has to calculate, yet watch how complex it needs to be. And then consider how even the most stupid animal has a mind blowing ability to do pattern recognition while we need to do complex CAPTCHA algorithms. I think that's an example of the fields not competing, but actually us lacking research in one of them. An exciting future ahead for sure. - SomaSynth, on 10/12/2007, -0/+4You may be over-complicating the problem. You're presuming that it's necessary for all molecular processes to be simulated, this is obviously not the case, it's only necessary to model the end-responses of all those chemical processes, that is, the information patterns (electrical responses) they stimulate. The distribution of these neurochemicals could easily be simulated without tapping into their atomic interactions, much like we can simulate aerodynamic flow without simulating the molecular constituents that create it. More than this, we might not even find it desireable to exactly model the physical distribution of these neurochemicals, or even precisely model electrical responses. Once the mind is free from the physical world, there's no reason for it to continue being limited by an artificial model of that world. We will choose not what is closest to the physical world but what works best, and the best solutions tend to be the simplest ones.
- brianeisley, on 10/12/2007, -1/+5@FizixMan: You're right that it comes down to logical instructions (true, false, Booleans, etc.), but what about neural nets? You can take logic gates and connect them together to create a device that will process input in the same fuzzy way that brains work.
Not that there's not probably a whole lot that this simulation is missing. But, still, it's a big step forward. What I want to know is, when they feed it input, what kind of output do they get? Is it making decisions based on things in its "environment"? And does it respond to changes to that input? If so, then that's the real story, because that means it's doing something that resembles thinking. - lowerlogic, on 10/12/2007, -0/+4Imagine once we can simulate a human mind in real time - without a body to care about, the AI could learn stuff 24/7. With access to the correct knowledge databases, it could learn everything about computers and learn how to make a faster computer. Once running in the new computer it could learn how to make an even faster one in less time. Singularity here we come!
- thirdoffive, on 10/12/2007, -0/+3True, this is going to be a major point of contention over the next century. I say if it has the same cascade of actions and reactions that a biological brain has, then there is no reason not to think that it is a being undergoing the same experience that a being supported by an equivalent biological brain would be going through.
- smackhero, on 10/12/2007, -0/+3what does tetrapedal-cybernetic-cheese taste like?
- bIuebonics, on 10/12/2007, -3/+6@FizixMan
is it then not interesting that quantum mechanics deals with discreet variables? - imikedaman, on 10/12/2007, -1/+4"//sarcasim"
Thanks for pointing that out.
Anyway, even if they did create Pinky, we all know he'd ultimately fail at every last plan to take over the world. - thirdoffive, on 10/12/2007, -0/+3Instead of programming a hooker you could program a girl who actually likes you (and is bi-curious).
- SundayBrunch, on 10/12/2007, -0/+3I would so simulate me a hooker.
- AussieScribe, on 10/12/2007, -0/+3> US researchers have simulated half a VIRTUAL mouse brain
No they haven't! They've simulated half a REAL brain.
Sheesh. Journalism today! :-(
Tim North
Perth, Western Australia - smackhero, on 10/12/2007, -0/+3@Logikos: that's a very interesting notion to think about--a world where materialism, superficiality, genetic predispositions, etc. have no hold. i think if such a matrix-type cyber universe really were created it would develop a very different culture from ours.
- ZPWeeks, on 10/12/2007, -0/+3Half a mouse brain on IBM BlueGene?
That means that you could simulate Paris Hilton's brain on a mac mini, right? - SomaSynth, on 10/12/2007, -0/+3We won't be creating digital humans for any experimentational purposes I'm sure, much like we don't and won't grow laboratory humans for drug testing.
When sufficient technology is available to replicate your own 'neural network', perhaps through high resolution external scanning, you could voluntarily choose to become digital. The more advanced the digital environment becomes, the more desireable that may be, but your humans rights would be intact as much as they are now. At least I'd like to think so, I find it hard to believe the world would allow for mass torture and slavery because it's a non-physical being. - Mousse, on 10/12/2007, -3/+6"No matter how powerful a super computer is, it really comes down a series of logical instructions. True, false, or, and... etc.
Which, when you think about it, is pretty dumb. Imagine describing life that way."
It's not so dumb once you realize you can represent fractions and other points on a continuum this way. It's a bit misleading, really, since binary is really just a numerical system that uses a base of two.
It would be like saying that mathematics is a dumb model because we only use 10 "states", the numbers 0 to 9, where in reality mathematics helps us define and explain the bulk of reality with beautiful models that operate on both discrete sets and continuums. - wush, on 10/12/2007, -1/+3moused a computer on a brain
- GeneralKickass, on 10/12/2007, -0/+2Bluegene-Shmoogene. Lets see that computer run my hamster wheel.
- thirdoffive, on 10/12/2007, -0/+2This article is confusing: “Half a real mouse brain is thought to have about eight million neurons” and then it says “Using this machine the researchers created half a virtual mouse brain that had 8,000 neurons.” Since when is 8 million the same as 8 thousand???
- Visk, on 10/12/2007, -0/+2I'm pretty sure this isn't the world's first computer mouse...
- DeusNova, on 10/12/2007, -1/+3Interesting, hopefully they'll perfect it. If this becomes huge, I think there will be a moral issue. Personally I'm for it if we can successfully apply it to human brains. We can have some of the smartest people preserve their brain and emulate it onto a computer, that would be so cool.
- ropers, on 10/12/2007, -1/+3@trainingname:
I can never understand the science fiction concept of a superior artificial intelligence threatening to wipe out humanity.
I mean. If said lifeform were any smarter than us, it would certainly know that exterminating other species is a really dumb idea.
Also, trainingname, if you think that sentinent AIs will be interested in killing other, possibly less developed sentinent beings off, well, I hate to ask this, but what does this say about YOU? - inactive, on 10/12/2007, -0/+2then steve jobs will put a lock on them all and sell out to the digital brain management crowd
- Ratteler, on 10/12/2007, -2/+4George Bush has a brain?
Hell... if we network all the Palms that people no longer use... we can emulate the cognitive power of the entire Christian right. -
Show 51 - 95 of 95 discussions
The Digg Toolbar for Firefox lets you Digg, submit content, and keep track of Digg even when you're not on the Digg site. Download the official 
© Digg Inc. 2009
— Content created and posted by Digg users is