Million Processor Network Models The Human Brain
If you thought the human brain couldn’t be modeled, think again. The human brain contains 100 billion neurons, each neuron can connect with up to 1 billion other neurons. So how can we simulate the human brain?
A group of researchers from British Universities are working on developing a massive supercomputer that makes use of up to a million ARM processors. The machine will be called SpiNNaker (Spiking Neural Network Architecture) and will be used to simulate the inner workings of the brain.
The project will no doubt further research in cognitive science, where scientists work on advancing our understanding of artificial intelligence. In the SpiNNaker model, each neuronal impulse (action potential) is modeled with a packet of data.
The neurons themselves are represented by a small simple equation that is solved in real-time through the individual ARM processor. Although we are not sure how much power will be required to power the millions of computer chips, you can bet it will be expensive.
The simulated neural network will help us understanding how the brain works, and although it does not involve the important variable of neurochemicals, it does help us understand how the brain think, although it doesn’t explain the influence that emotions have on the brain.