The NZ$1 million-dollar project at the University of Canterbury marks one of the latest developments in ‘neurocomputing’, in which machines and computers are designed to mimic the functionality of human brains – specifically, the ability to learn. Our brains contain neurons that are essentially connected by synapses, and it is the behaviour of these synapses that influences our learning.
So says Professor Simon Brown, who is leading the study with colleague Dr. Saurabh Bose.
“When you associate a particular image with a particular idea, it’s because of the pathway through the synapses which has been created,” he said.
The scientists’ ambitious approach to neurocomputing with microchips sees them creating networks made up of ‘nano-scale junctions’ that perform as ‘tiny switches’. This network is designed to form connections, or pathways, as synapses do in the brain.
“On a simplistic level, the switches behave in a similar way to the synapses,” said Brown.
What the chip’s ultimate capability will be remains uncertain.
“Can we actually teach these networks to do calculations?” Brown added. “Or can we teach them to do pattern recognition, which is one of the standard things that people think these networks should be good at, because this is one of the things that the brain is really good at.
“It’s likely that it will perform some tasks like pattern recognition much better than an ordinary computer.”
By the end of the study, which is set to take two and a half years, Brown hopes to have a new company specialising in the creation of neomorphic computer chips.
And one of the additional crucial benefits of neurocomputing transferred to a microchip must be that the energy required to power it is minimal.
“One of the big advantages of moving from a supercomputer to a chip is that the chip should be able to do tasks with 100 times less power consumption.
“Of course, if you’re using less power, that’s what you want for mobile devices like drones and cellphones.
“We think this is a big advance because our new chips work for months and months, whereas previous chips of this type only lasted for hours.”
Scientists in New Zealand have commenced work on a study that could see the capabilities of the human brain compacted into a microchip that is small enough to fit in a smartphone.