Print 22 comment(s) - last by AnnihilatorX.. on Nov 26 at 6:00 AM

  (Source: ICanHasCheezBurger)
Leading neuroscientist accuses government-funded researcher of deception

The Defense Advanced Research Projects Agency (DARPA), the premier funding branch of the Pentagon, has poured $5M USD into a project with IBM to make electronics that mimic the "size, function, and power consumption" of a cat brain.  And last week, it appeared that a major breakthrough had been made with IBM’s lead researcher on the project, Dharmendra Modha, claiming to have simulated all the neurons and synapses of the cat brain on a supercomputer.

Now the accuracy of that claim and the validity of the project is being called into question.  Henry Markham, a leading neuroscientist blasted the project and Mr. Modha's claims calling them a "hoax".  Ironically, Mr. Markham also works on a project affiliated with IBM, dubbed "Blue Brain".  He complains that Mr. Modha simply put together a "PR stunt here to ride on Blue Brain.”

He describes in an open letter to Bernard Myerson, IBM’s Chief Technology Officer, "What IBM reported is a scam - no where near a cat-scale brain simulation.  I am absolutely shocked at this announcement. Not because it is any kind of technical feat, but because of the mass deception of the public."

He continues:

All these kinds of simulations are trivial and have been around for decades - simply called artificial neural network (ANN) simulations. We even stooped to doing these kinds of simulations as bench mark tests 4 years ago with 10’s of millions of such points… If we (or anyone else) wanted to we could easily do this for a billion “points”, but we would certainly not call it a cat-scale simulation. It is really no big deal to simulate a billion points interacting if you have a big enough computer. The only step here is that they have at their disposal a big computer. For a grown up “researcher” to get excited because one can simulate billions of points interacting is ludicrous.
This is light years away from a cat brain, not even close to an ants brain in complexity. It is highly unethical of Mohda to mislead the public in making people believe they have actually simulated a cat’s brain…. That IBM and DARPA would support such deceptive announcements is even more shocking.

Mr. Modha had claimed last week that his simulation of a billion neurons and 10 trillion neural synapses on IBM hardware was roughly the equivalent of a cat brain.  His colleagues lauded the accomplishment, with some comparing his work to the magnitude of work done on the Large Hadron Collider project.

Mr. Myerson's criticism is diverse.  Among other things he says that the point representation of neurons are incomplete as they don't simulate the complex effects of ion channels or more complex phenomena like neuronal branching. 

He says the "brain" developed isn't even as complex as an ant brain and is hardly praise worthy.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

it's different
By mahax on 11/24/2009 10:58:05 AM , Rating: 2
animal brain is a sum of all it's input and output (action, reactions). We are defined by our lives and if you just set up these neurons in software, what will they do? Interact randomly? Animal brain starts it's growth from only a couple of (I presume) specialized cells which then form up by some plan or logic. Setting up billion of them just like that would, in my mind, seem to equal only braindead. The software would have to be able to "grow" the brain first...

RE: it's different
By grath on 11/24/2009 12:12:21 PM , Rating: 2
It would be braindead in the sense of a CPU not being plugged into a motherboard. The point is to not just simulate the brain, but also to simulate the inputs and outputs you speak of and how the brain processes them. Its better to start with a developed adult brain, determine some basic I/O functionality, then work incrementally backwards from there toward a less developed.

You are probably right, that to truly simulate complex emergent behavior from a biological system would require the system to develop under I/O exposure from the early stages, but it is better to reverse engineer the finished product than to start with a couple brain stem cells and try to recreate evolution.

"Game reviewers fought each other to write the most glowing coverage possible for the powerhouse Sony, MS systems. Reviewers flipped coins to see who would review the Nintendo Wii. The losers got stuck with the job." -- Andy Marken

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki