RESEARCHERS AT MIT have attempted to design computer chips that work like the human brain (hopefully, the human brain of someone relatively normal).
Dubbing the study part of the emerging field of “neuromorphic computing”, a team of engineers at the university have apparently designed an “artificial synapse” that means they can precisely control the strength of an electric current flowing across it, similar to the way ions flow between neurons in the brain.
“There are more than 100 trillion synapses that mediate neuron signalling in the brain, strengthening some connections while pruning others, in a process that enables the brain to recognise patterns, remember facts, and carry out other learning tasks, at lightning speeds,” claimed MIT in a news post.
“Instead of carrying out computations based on binary, on/off signalling, like digital chips do today, the elements of a ‘brain on a chip’ would work in an analogue fashion, exchanging a gradient of signals, or ‘weights’, much like neurons that activate in various ways depending on the type and number of ions that flow across a synapse.”
Before starting the experiement, MIT thought that, in theory, this would mean small neuromorphic chips could, like the brain, efficiently process millions of streams of parallel computations that are currently only possible with large banks of supercomputers.
The researchers therefore fabricated a neuromorphic chip consisting of these artificial synapses made from silICOn germanium, each synapse measuring about 25 nanometers across. They applied voltage to each synapse and found that all synapses exhibited more or less the same current, or flow of ions, with about a 4 per cent variation between synapses. This was a much more uniform performance compared with synapses made from the alternative, amorphous material, of which many other chis have been made.
The team then tested a single synapse over multiple trials, applying the same voltage over 700 cycles, and found the synapse exhibited the same current, with just 1 per cent variation from cycle to cycle.
“This is the most uniform device we could achieve, which is the key to demonstrating artificial neural networks,” said team leader Jeehwan Kim, the Class of 1947 Career Development Assistant Professor in the departments of Mechanical Engineering and Materials Science and Engineering.
In simulations, Kim and his the researchers therefore found that the chip and its synapses could be used to recognise samples of handwriting, with 95 per cent accuracy.
The study was published today in the journal Nature Materials. MIT said the development is a major step toward building portable, low-power neuromorphic chips for use in pattern recognition and other learning tasks.
The team is now in the process of fabricating a working neuromorphic chip that can carry out handwriting-recognition tasks, not in simulation but in reality.
“Ultimately we want a chip as big as a fingernail to replace one big supercomputer,” Kim says. “This opens a stepping stone to produce real artificial hardware.” µ
Source : Inquirer