Neural networks were once heralded as the future of computers, because of their ability to learn rather than require manual programming. However it was quickly discovered that application-specific designs would not yield general purpose solutions. Now IBM is aiming to solve that problem by emulating widely available magnetic resonance diffusion weighted imaging data (shown by colors in photo). Look for brain-accurate simulations to spawn a new era of cognitive computers within 10 years. R.C.J.
IBM claims to have succeeded in creating a supercomputer simulation of the human brain capable of sensation, perception, action, interaction and cognition. So far it can only emulate the number of neurons and synapses in a cat's brain—1 billion spiking neurons and 10 trillion individual learning synapses— but then again the project has only been underway for one year.
IBM's paper on the project, called "The Cat is Out of the Bag" was awarded the Association for Computing Machinery (ACM) Gordon Bell Prize at the recent Supercomputer 2009 conference.