Future Computers May Replicate Human Brain Synapses

While the development of artificial intelligence is inspired by the replicating the learning and problem-solving capacities of the human brain, the computer systems that they run on bear little resemblance to our grey matter. Researchers from Stanford University and Sandia National Laboratories, though, are one step closer to recreating the intellectual hardware encapsulated by our skulls. In a paper published this week by Nature Materials, the researchers reveal that they have created an artificial synapse, the structure in the brain that allows electrochemical signals to jump between cells.

“It works like a real synapse but it’s an organic electronic device that can be engineered,” said Alberto Salleo, Associate Professor of Materials Science and Engineering at Stanford and senior author of the paper, told VR-Zone. “It’s an entirely new family of devices because this type of architecture has not been shown before. For many key metrics, it also performs better than anything that’s been done before with inorganics.”

“Deep learning algorithms are very powerful but they rely on processors to calculate and simulate the electrical states and store them somewhere else, which is inefficient in terms of energy and time,” Yoeri van de Burgt, lead author of the paper, added. “Instead of simulating a neural network, our work is trying to make a neural network.”

The artificial synapse is made of two metal films, untouching, connected by an electrolyte. It has been tested as part of an AI system designed to analyse handwriting – recognising numerical characters between 0 and 9 – during which it boasted a success rate of 93-97%.