Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Biotech Science Technology

Flying By Brain 636

Posted by timothy from the brain-is-my-co-pilot dept.
Garabito writes "Scientists at the University of Florida made a living 'brain' by extracting 25,000 neurons from a rat's brain and culturing them inside a glass dish. Then, the neurons began to extend lines to each other, creating a living neural network between them. The dish had a grid of 60 electrodes connected to a computer running a flight simulator. The scientists were able to train the 'brain' to control the plane in the simulator and to react to conditions of the plane. Are we getting closer to create an artificially made conscious being, or perhaps, a living computer?" AlphaJoe was one of several readers to add a link to Wired's article on the experiment.
This discussion has been archived. No new comments can be posted.

Flying By Brain More

Flying By Brain

Comments Filter:

  • working backwards (Score:3, Insightful)

    by man_ls ( 248470 ) on Saturday October 23, 2004 @08:01PM (#10611572)
    We designed neural networks to follow how brains work.

    Now we're using a brain to run a neural network.

    Chicken-egg problem, anyone? :)

  • rat brains (Score:4, Funny)

    by Anonymous Coward on Saturday October 23, 2004 @08:02PM (#10611577)
    the last thing i want is a rat flying my plane

  • Uhm, not the appropriate response, but (Score:5, Funny)

    by MrHanky ( 141717 ) on Saturday October 23, 2004 @08:02PM (#10611578) Homepage Journal
    The first thing I thought was: I want one. Wonder if it could learn to play GTA?

  • Abby someone (Score:5, Funny)

    by Alien Being ( 18488 ) on Saturday October 23, 2004 @08:03PM (#10611584)
    Dr. Frederick Frankenstein : Igor, would you mind telling me whose brain I did put in?
    Igor : And you won't be angry?
    Dr. Frederick Frankenstein : I will NOT be angry.
    Igor : Abby someone.
    Dr. Frederick Frankenstein : Abby someone. Abby who?
    Igor : Abby Normal.
    Dr. Frederick Frankenstein : Abby Normal?
    Igor : I'm almost sure that was the name.

    • Re:Abby someone (Score:5, Informative)

      by pipingguy ( 566974 ) on Saturday October 23, 2004 @11:52PM (#10612576)

      What hump?

      The quoted dialogue above is a hilarious exchange from an extremely funny movie [imdb.com]. They made it in B&W and it still worked in 1974. Today it's quite a cult classic.

      I haven't seen a slashdot name of Abby Normal yet and you can always slip brains through slot in door after 5PM.

  • What's next.. (Score:5, Funny)

    by KinkifyTheNation ( 823618 ) on Saturday October 23, 2004 @08:05PM (#10611592) Journal
    Are we getting closer to create an artificially made conscious being, or perhaps, a living computer?
    Or better yet, self-controlled flying lawnmowers!

  • Brain bags! (Score:3, Funny)

    by jfarnold ( 320693 ) on Saturday October 23, 2004 @08:05PM (#10611596)
    Soon we will all be augmented by our extra brain bags! Organic computers in a purse that we either wear or have implanted in our abdomens. I can't wait for the beta test.

    • Re:Brain bags! (Score:5, Funny)

      by RedCard ( 302122 ) on Saturday October 23, 2004 @08:30PM (#10611740)
      Soon we will all be augmented by our extra brain bags! Organic computers in a purse that we either wear or have implanted in our abdomens

      To quote the work of Scott Adams... [dilbert.com]

      Dogbert: (Talking to PHB at the office) The dogbert consulting company will plot a new course for your business

      Dogbert: My consultants are so smart that their brains don't fit in their heads. They have to strap the extra brains to their torsos.

      Ratbert: (Later at home) Why do I need a piece of liver strapped to my torso?

      Dogbert: I got a little carried away at the pitch meeting.

    • Re:Brain bags! (Score:5, Funny)

      by node 3 ( 115640 ) on Saturday October 23, 2004 @08:49PM (#10611836)
      Soon we will all be augmented by our extra brain bags! Organic computers in a purse that we either wear or have implanted in our abdomens. I can't wait for the beta test.

      People with their brains implated in their (lower) abdomen? We've already got those. They're called "The RIAA".

  • Does this......? (Score:5, Insightful)

    by Anonymous Coward on Saturday October 23, 2004 @08:06PM (#10611599)
    Does this freak the shit out of anyone else?

  • ObSimpsons.... (Score:5, Funny)

    by Elminst ( 53259 ) on Saturday October 23, 2004 @08:07PM (#10611604) Homepage
    I for one welcome our new plane-flying rat-brain overlords...

  • Fire Ron Zook (Score:4, Funny)

    by daidojiuji ( 183833 ) on Saturday October 23, 2004 @08:08PM (#10611610) Homepage
    As a recent graduate of the University of Florida, I have one question to ask of these researchers: How many days do we have to wait until they have a prototype that can function as the football team's head coach? It can't be too hard to do better than Coach Zook.

  • One question... (Score:5, Interesting)

    by doublebackslash ( 702979 ) <doublebackslash@gmail.com> on Saturday October 23, 2004 @08:08PM (#10611613)
    How did the clump of neurons know what they were trying to accomplish? More precicely, why didn't they try to crash the plane? What sort of positive/negative feedback did they use? I understand that this works, and vaugely how it works, but i can't wrap my poor little brain around what sort of feedback they used!
    • Exactly what I wondered too, and I did even read the whole article to see if it was mentioned.

      I suppose that the goal would be to keep the plane level and heading straight ahead or something, then the brain learns how to accomplish this, thus allowing it to fly in different conditions. But I couldn't find any info on how the brain was told this was the "right" thing.

      Maybe they just let the simulator fly the plane straight ahead without interference until the brain learnt that this was "normal", then, when

    • Re:One question... (Score:4, Funny)

      by MrHanky ( 141717 ) on Saturday October 23, 2004 @08:18PM (#10611674) Homepage Journal
      i can't wrap my poor little brain around what sort of feedback they used!
      Obviously. If you did, you'd be flying a flight simulator, not posting to slashdot.

      In other news,[1] rats have made clumps of neurons from scientists' brains behave in a crude sort of stimulus-response behaviour by connecting the neurons to a simulation of a news for nerds site.

      [1]Or should that be 'In Soviet Russia...'?

    • Re:One question... (Score:5, Informative)

      by starm_ ( 573321 ) on Saturday October 23, 2004 @09:53PM (#10612129)
      Well I took a course about artificial neural net (not the biological ones like here). But we learn that biologocal neurons learn by repetition and correlation. When a neuron sees a pattern it tries to repeat it. They probably ran the simulator under different conditions. While giving input to the neurons they forced the output signals. (with simple voltages) The neurons learned these output signals. Afterwards, they just had to give the inputs signals and the neural net would automatically give the output signals it got used to.

      basically the net learns an unlinear function or the inputs. outputi = fi(input1,input2,input3,input4 ...) and these are all voltage pulses (caused by chemical reactions and input signal from the computer)in the neural net.

    • Re:One question... (Score:5, Informative)

      by Quantum Jim ( 610382 ) <jfcst24NO@SPAMyahoo.com> on Saturday October 23, 2004 @11:02PM (#10612402) Homepage Journal

      why didn't they try to crash the plane? What sort of positive/negative feedback did they use?

      The second article stated that neurons were given information on the tilt of the airplane:

      To control the simulated aircraft, the neurons first receive information from the computer about flight conditions: whether the plane is flying straight and level or is tilted to the left or to the right. The neurons then analyze the data and respond by sending signals to the plane's controls. Those signals alter the flight path and new information is sent to the neurons, creating a feedback system.

      It seems that this experiment builds on earier research by DeMarse, Wagenaar, Blau, and Potter in 2001 called the the animat [gatech.edu]. It wondered in a box without goal-specific behavior. However, it also tended to specific patterns and states. That is a very readable article - I highly suggest you read it.

      But why did the neurons want to stablize the aircraft? I couldn't find a paper on the aircraft experiment, but a second paper, "Removing some 'A' from AI: Embodied Cultured Networks" (by Bakkum, Shkolnik, Ben-Ary, Gamblen, DeMarse, and Potter, 2004) summarized another experiment where neurons were trained to keep a set distance from an object. The paper is the first article on the same page [gatech.edu] of publications as the first paper. It seems that the neural network responded nonlinearly - that is, it changed state from one behavior to another one - when the input stimulus frequency was adjusted (correct me if I'm wrong). So by changing the input stimulus frequency, they were able to train the network. I gather that the new experiment simply uses when certain "level = good, nonlevel = bad" stimuli. It's a long way off from Robocop II, but it is a start.

      • Re:One question... (Score:5, Interesting)

        by Illserve ( 56215 ) on Sunday October 24, 2004 @06:26AM (#10613436)
        Someone correct me if I'm wrong, but it's hard to tell because these guys specifically avoid using the technical terminology of the LTP (Long Term Potentiation) literature, probably because they know they aren't getting it and don't want to step into that minefield.

        As near as I can tell from their paper at:
        http://www.neuro.gatech.edu/groups/potter/papers/D agstuhlAIBakkumpreprint.pdf

        the network is not "learning". Rather, they are setting up the system so that the inherent properties of the neurons cause the correct response to the feedback it receives from the environment.

        The real knowledge about the task is built into the systems that interface with the neurons.

        As an analogy, the neuron is behaving like a spring in a mechanical system, it has some basic fundamental properties that are statistically predictable, and the system around the spring expects it to behave thusly. But because it's a complex system it may take time for the system to settle into the stable state, hence it looks as if the network "learns", when really it's a system of springs settling into an equilibrium.

        Not to understate their technical accomplishments. They've done amazing things with cultured neurons. But this is not about reward and punishment, the network is far too simple for such words to have any meaning. It may not even be about learning in the sense of permanently modifying synaptic connections. I can't tell from my first read through, and that's what really sets off the alarm bells.

        They also avoid the obvious experiment that should be done if they think long term plasticity is involved. (ie, can it still navigate the next day?)

        • Re:One question... (Score:5, Informative)

          by NeuroHx0r ( 777571 ) on Sunday October 24, 2004 @11:09AM (#10614281)
          We did not report LTP because it is NOT LTP. In fact, we are using and effect reported by Eytan, D., Brenner, N., and Marom, S., Selective Adaptation in Networks of Cortical Neurons. Journal of Neuroscience, 2003. 23(28): p. 9349-9356 in which "high" frequency stimulations (once every second) was reported to depress the response of the network while "low" frequency stimulations resulted in an enhanced response. For our system we tied the network's response to the control surfaces, dedicating stimulations on one channel for pitch, and a second for roll control. Each channel is stimulated separately, and the response (PSTH) is recorded. Control movements are proportional to the current error from straight and level by mapping the error (0 to 180 degrees) to the interval 0 to 100 ms of the PSTH and integrating the difference in response before training, to the current or enhanced or depressed levels. The more error, the more the control surface is moved. The networks only gradually control the aircraft since the Marom effect requires over 15 minutes to develop. The two frequencies are then used to adjust these weights (i.e. number of spikes in the PSTH) to produce optimal flight. The neurons/network don't seek optimal flight in the classic sense. Instead, we adjust the weights (using high and low Freq. stims) in the network to produce that result. It is a very simple system and our only interest in it is in terms of those changes within the network and the possibility to extend it to more of the network than just two or three different channels. Hope that helps.. Tom DeMarse

    • Re:One question... (Score:5, Interesting)

      by tyler_larson ( 558763 ) on Sunday October 24, 2004 @12:28AM (#10612674) Homepage
      How did the clump of neurons know what they were trying to accomplish? More precicely, why didn't they try to crash the plane?

      I think it's significant that they chose a flight simulator instead of a more traditional "game" to teach the newly formed brain.

      Here's a couple of points to remember:

      The difference between the makeup, function, and behavior of a given type of cells between one species and another is so insignificant (remember, we're talking on a cellular level) that they can generally be ignored. You can almost always assume that a given cell type in one organism will behave identically to a parallel cell in another. The species that the cell came from is all but insignificant.

      Brain cells, (in humans and in other species) are amazingly versatile. While capable of specializing (vision centers, speech centers, etc.), these cells seem to be capable of taking on any function necessary for the benefit of the organism. For example, humans brains in which a specific part has been damaged (such as the vision center) have actually re-mapped other cell groups to take over that function. They do what they have to to survive.

      Brain cells are cooperative in nature: if placed in proximity to eachother, they'll work together for their common good (read: survival). They'll "instinctively" form a structure similar to how they're pre-designed to work. They'll form a brain--as fully functional as the situation permits. It doesn't necessarily matter how you arrange them, the brain cells can sort those details out--somehow.

      Brains look for order. We've known that for ages. Finding order is how a brain learns, it's how the brain separates relevant details from the background noise. The ability to identify order is the whole basis of intelligence. Every sense, every stimulus, every aspect of the brain has order-seeking overtones. This feature of brains is so absolutely universal that it must be deeply ingrained into the neurons themselves.

      Put those details together, and you end up with the following scenario: if you take neurons out of an organism and place them together, they'll form a brain. Probably not as complex or capable a brain as you started with, but a brain none the less. Actually this is the ideal brain to study, as you're starting "from scratch": there's no evolutionary specialization involved. Each cell will attempt to make sense of its neighbors, and as a result, the organism as a whole will attempt to make sense of its environment (brain processes are the ultimate in emergent algorithms). The brain will follow this behavior as if it were necessary to the brain's survival.

      Which brings us to the flight simulator. If you instead had the brain play with a chessboard or a clock, the results would probably be unimpressive. But a flight simulator--that's really the perfect environment. There's the potential for the brain to actually order its environment: there are equilibrium points that the brain will eventually find where it has greater control over its inputs. Assuming that flying too hight or too low creates a more chaotic state, you can likely expect the brain to learn to avoid it.

      In fact, I'd be very much surprised if you didn't actually see the brain cells start to specialize. Some cells will become responsibe for directly manipulating the flight controls based on the inputs from the brain. Some will attempt to maintain aircraft equilibrium in absence of any other input from the brain. Others will control the aircraft as a whole, their location in the network giving them a better overall picture of the situation than, say, the cells near the controls. Furthermore, I fully expect some cells to not participate at all: cells that are "out of the loop", so to speak, will proably cease most activity to avoid disturbing the overall process.

      I, personally, have been waiting to see this very experiment conducted and see the results. I think this is very exciting science.

  • obligatory (Score:5, Funny)

    by Anonymous Coward on Saturday October 23, 2004 @08:08PM (#10611614)

    (in drone-like monotone)
    Imagine a beowulf cluster of these things.

  • Great now im going to lose my job (Score:5, Funny)

    by XST1 ( 824817 ) on Saturday October 23, 2004 @08:09PM (#10611616)
    As an airline pilot for American, its nice to see my job being outsourced by rats in the future.

    • Re:Great now im going to lose my job (Score:5, Funny)

      by GileadGreene ( 539584 ) on Saturday October 23, 2004 @09:06PM (#10611932) Homepage
      Reminds me of a classic aviation joke one of my commerical pilot friends told me once:
      Back in the day, a big plane took a crew of 5 - pilot, copilot, navigator, flight engineer, and radio operator.

      Then radio technology improved, and they eliminated the radio operator, so it was down to 4.

      Next to go was the navigator, as long range navigation beacons became prevalent. So we're down to 3 crew members.

      And even those days are numbered - as planes have become more computerized, flight engineers have become unnecessary, and many newer planes don't require them. So in a lot of cases we're down to just 2 crew members, pilot and copilot.

      My friend truly believes that the next step in aviation automation is to eliminate the copilot. Instead, the crew will consist of a pilot, and a dog. The pilot is there to feed the dog...



      ...and the dog is there to bite the pilot if he touches anything in the cockpit. ;)

  • Great, how long until... (Score:4, Funny)

    by tao_of_biology ( 666898 ) <tao,of,biology&gmail,com> on Saturday October 23, 2004 @08:09PM (#10611617)
    they outsource my programming job to a petri dish...

  • Living 'eh? (Score:3, Interesting)

    by macaulay805 ( 823467 ) on Saturday October 23, 2004 @08:10PM (#10611627) Homepage Journal
    Bring a whole new meaning of a computer virus ...

  • Do you have to think in Russian? (Score:5, Funny)

    by Omega1045 ( 584264 ) on Saturday October 23, 2004 @08:11PM (#10611633)
    Do you have to think in Russian?
  • So, how is this thing reacting to good and bad?

    Did they create a neural net that falls through a given search space to a local or global minimum, or what?

    Is "good" a total lack of input, i.e. the plane is flying straight with no lateral or vertical drift, and is degree of input dependent on the amount of lateral motion, etc.?

    As I type this, it makes sense that this might be so, but I wonder why the network created a negative feedback system, and not a positive feedback system.

    ~ Mike

  • Human neurons... (Score:5, Interesting)

    by zors ( 665805 ) on Saturday October 23, 2004 @08:14PM (#10611652)
    I wonder if human neurons would be more effective? Or are all neurons created equal, and only the structure of a brain makes it more or less intelligent? Could we grow rat neurons into a human brain? Maybe we could customize brains for certain abilities, by growing them along certain structures. I don't have alot of personal knowledge here, so i'm just putting out some questions that this brought up for me.

  • sea slugs (Score:4, Interesting)

    by Anonymous Coward on Saturday October 23, 2004 @08:27PM (#10611724)
    I know of similar work with sea slugs in an off-campus lab funded partly by UF (the Whitney Lab). I'm not _too_ familiar with it, so this may not be entirely accurate. Basically, they found that neurons in the brain of the seaslugs are always in the same positions as other animals of the same species. They then started training animals, much like pavlov's dogs, to close their siphon whenever they were electrically shocked on their tail (by touching the siphon whenever they were shocked so the animal would relate the 2 stimuli). They then could isolate the neurons in the brain and train then individually. Two neurons in a petri dish would gradually connect and then share information. At the moment the group is working on identifying which genes control what part of the brain, or something like that..

  • This is your brain... (Score:5, Funny)

    by Eric Giguere ( 42863 ) on Saturday October 23, 2004 @08:36PM (#10611771) Homepage Journal

    Adds a whole new dimension to the commercial, doesn't it?

    This is your brain...
    This is your brain on drugs...
    This is your brain on drugs flying a plane without you...

    Eric
    Why Vioxx is Prozac for lawyers [ericgiguere.com]

  • Pretty neat, but (Score:3, Informative)

    by slobber ( 685169 ) on Saturday October 23, 2004 @08:37PM (#10611776)
    "If you think about your brain, and learning and the memory process, I can ask you questions about when you were 5 years old and you can retrieve information. That's a tremendous capacity for memory.

    I have to say, I don't remember much from when I was five years old. I remember where I lived and maybe can guesstimate where I spent a specific summer, but most of my knowledge comes from what my parents told me and from little "text" snippets that somehow got stuck in my head (for example, names of cities I visited, etc.)

    I can recall some images from the past, but I am not sure whether those are "true" memories or something synthesised by brain to "fill in the blank". This leads me to believe that human memory is rather lossy and large part of what I remember is just a rough approximation of what happened based on a few datapoints that brain actually remembers. Sort of like with people who have a defect in their iris - they still see an image in what's supposed to be a blind spot. This image is synthesised by brain to fill in the gap. Needless to say, occasionaly it turns deadly (especially while driving).

  • Ladies and gentlemen, from the flight deck (Score:5, Funny)

    by HangingChad ( 677530 ) on Saturday October 23, 2004 @08:41PM (#10611793) Homepage
    This is your captain, Rat Brain 4023, integrated neural network and my first officer, Rat Brain 4024. We'll be flying at an altitude of 30,000 feet and are expecting a nice smooth ride-- HOLY SHIT CHEESE!!! LOOK OVER THERE IT'S CHEESE!!! Ooop, sorry about that, false alarm. We're expecting nice weather in HEY THERE"s A F*ING CAT IN THE CARGO HOLD!!! Eject! Eject! Eject!

  • Saw this a few days ago... (Score:5, Funny)

    by jafiwam ( 310805 ) on Saturday October 23, 2004 @09:12PM (#10611960) Homepage Journal
    ... and something is REALLY bugging me about it.

    How do you motivate a slice of rat brain to fly a plane? Does it feal pain when it crashes? Get nutrients when it flys far? What?

    All too soon we will see little USB plug ins with these things to help the rail-gun spawn-campers aim fast in UT2024; Ultimate.

    [FuZZy1] Punched a hole in 3L1T3's cranium
    [3L1T3>] NOOB!
    [3L1T3]; Rat-bot camper!
    [FuZZy1]; LOL!1 That why tehy call me Fuzzy1
    • My guess is, you use the electrodes in the petri dish to stimulate the neurons into strengthening the connections when the plane is doing the right thing (straight and level presumadly) and weaken (or just not strengthen) when the plane is doing the wrong thing (crashing).

      I believe that by altering the characteristics of the charge applied over the electrodes this effect could be realised.

      Eventually the connections will be strengthened in such as way as the plane is flown straight and level.

      Nothing to do

  • This is an old story. (Score:4, Funny)

    by node 3 ( 115640 ) on Saturday October 23, 2004 @09:17PM (#10611973)
    Rat brains flew a plane for the National Guard to get out of the Vietnam War.

  • The rodent-feline arms race has begun. (Score:5, Funny)

    by PinchDuck ( 199974 ) on Saturday October 23, 2004 @09:42PM (#10612081)
    Rats are ugly and disgusting and already have claws and teeth and biological weapons capability...now we give them Sidewinders, air-to-ground missles and 20 MM cannon. That's disturbing.

    I'm immediately going to deploy a network of cat-neuron controlled anti-aircraft missle batteries.

    damned rats.

  • Different brain cells (Score:5, Funny)

    by Tablizer ( 95088 ) on Saturday October 23, 2004 @09:43PM (#10612083) Journal
    They tried brain cells from different individuals. Here is the result:

    Osama's cells: Plane kept crashing into buildings.

    PHB cells: Plane kept flying in circles until it ran out of gas.

    Bill Gates cells: Plane kept locking up.

    SCO lawyer cells: Plane kept crashing, but blaming other planes.

    RMS cells: Plane wanted to call itself "GNU Plane".

    G.W. Bush cells: Plane kept crashing into Saddam Hussein no matter what, even if Osama was placed right next to Saddam.

    John Kerry cells: Plane would fly to the left, and then to the right, and then to the left....

    Slashdot reader cells: Plane would try to fly without first reading the flying manual.

    Steve Jobs cells: Plane transformed itself into a slick, modern, translucent jet, but priced itself too high.

    Mike Melvill cells: Plane kept going up and up until we lost track of it.

    Emacs coder cells: Plane became a boat, a car, a house, a lawn mower, and a finger-nail clipper.

  • I can't believe it (Score:5, Funny)

    by blamanj ( 253811 ) on Saturday October 23, 2004 @10:00PM (#10612153)
    Nearly 200 responses and nobody has asked if it runs Linux.

  • There is not a single cyberpunk or SF writer who (Score:3, Funny)

    by multiplexo ( 27356 ) * on Saturday October 23, 2004 @11:00PM (#10612394) Journal
    could have come up with anything better than the first line of the article. As good as, yes, but better? No.

    Somewhere in Florida, 25,000 disembodied rat neurons are thinking about flying an F-22.

    It's just such a great hook.

Slashdot Top Deals

Solutions are obvious if one only has the optical power to observe them over the horizon. -- K.A. Arsdall

Close