IBM scientists have created artificial neurons and synapses using phase change memory (PCM) that mimics the brain’s cognitive learning capability.
For the first time in history, artificial phase-change neurons have been grouped together (in a population of 500 synthesized in a lab) to process a neurological signal in more or less the same way that biological neurons transmit messages. They can be made exceptionally small and are similar in power and energy usage to biological neurons, and can even produce results with random variations, also just like biological neurons.
For non-scientists, the importance of this discovery may not be immediately apparent. IBM’s artificial neuron, developed by a research team in Zurich, is quite literally the next best thing to a naturally created biological neuron. The lab-created version has all the same components of a biological neuron, including inputs (dendrites), a neuronal membrane (lipid bilayer) around the spike generator (soma, nucleus), and an output (axon). Likewise, its functions mimic those of its biological counterpart.
In addition to all that, the artificial neurons are durable, made from well-known materials that can withstand trillions of switching cycles. They are tiny (around 90 nanometers) and researchers believe they can make them even smaller, possibly as minuscule as 14nm. The researchers started by creating 500 artificial neurons together in a chain capable of sending signals, which means the IBM team has created the closest artificial version of a biological neuron. In the next phase of research, the team will create a much larger population of artificial neurons, with thousands of individual units, and write software to push their capabilities to the limit.
The study results were published in the journal Nature Nanotechnology.
Jean G. Thomas