Researchers from McGill University have developed a light-detecting nanoscale structure that mimics how a neuron process information. These single artificial neurons, created with light-detecting nanowires, demonstrate the ability to mimic many biological neuronal behaviors, including excitation and inhibition, threshold, refractory period, and temporal summation.
The neuron-like behaviors of the nanowire structure emerge directly from the semiconductor material from which it is built, rather than from software programming or complex circuitry. Instead of capturing data first and processing it elsewhere, the nanowire device senses and interprets light in the same place, similarly to how the eye processes visual information.
The new device, developed by a team led by professor Songrui Zhao, could boost the efficiency of vision-based technologies like artificial retinas and smart optical sensors. It could also affect how artificial neural networks (ANNs), a cornerstone of machine learning, are built. The single artificial neurons produced by the device could be used to build an ANN in a bottom-up manner.
“A single artificial neuron is like a cell you can use as a building block, allowing us to construct networks from the bottom up,” Zhao said. “It’s a bit of a crazy idea — to create something like a biological system using an inorganic material.”
To read more, click here.