MIT creates devices for artificial brain synapses

Engineers have put thousands of artificial brain synapses on a chip that is smaller than a piece of confetti, MIT News reported. 

The artificial brain synapses are known as memristors and are silicon-based components that are able to mimic the transmission synapses that are found in the human brain.

“So far, artificial synapse networks exist as software," Jeehwan Kim, associate professor of mechanical engineering at MIT, told MIT News. "We’re trying to build real neural network hardware for portable artificial intelligence systems."

Kim told the news agency they're hoping to use the memristors in real-time.

“Imagine connecting a neuromorphic device to a camera on your car and having it recognize lights and objects and make a decision immediately, without having to connect to the internet," Kim told MIT News. "We hope to use energy-efficient memristors to do those tasks on-site, in real-time."

The results of the research were published in the journal Nature Nanotechnology. The research shows that the brain-inspired circuits can be done on small devices that are portable and can carry out complex tasks that supercomputers can handle.

Kim also told MIT News existing memristor designs work well in certain cases, like where voltage stimulates a large conduction channel, but those designs are less reliable when memristors need to general signals that are more subtle and on thinner conduction channels.

Kim said the technique was metallurgy.

“Traditionally, metallurgists try to add different atoms into a bulk matrix to strengthen materials, and we thought, why not tweak the atomic interactions in our memristor, and add some alloying element to control the movement of ions in our medium,” Kim told MIT News.

Kim told MIT News the research team chose copper for the memristors because it could bind to both silver and silicon.

“It acts as a sort of bridge, and stabilizes the silver-silicon interface,” Kim told the news agency.

The research was funded by the MIT Research Support Committee funds, as well as IBM Watson AI Lab, the National Science Foundation and the Samsung Global Research Laboratory.