Atomic Switch Networks: A Biologically Inspired Approach to Neuromorphic Computation

Posted on

JamesGimzewski 3 0

Professor James Gimzewski was recently featued on the cover of the latest issue of Convergence, a magazine published by the International Center for Materials Nanoarchitectonics (MANA), Japan.

Atomic Switch Networks (ASN) are biologically inspired, self-organizing devices comprising of massively interconnected networks. MANA Satellite Director at UCLA, Professor James Gimzewski talks about how he discovered synaptic behavior of ASN and his path to put artificial brain into reality based on collaboration between MANA and UCLA. In between, he reveals how drones and yoga inspires him and how science has to develop in the future.

Inspired by a Brain

Vol.20 Gimzewski250*335
Professor James Gimzewski begins by saying, “There are many limitations to what a conventional computer can do.” However, he adds, “the human brain is able to manage and handle very complex situations even in a very noisy and error-prone environment.” In fact, scaling conventional computer to emulate the human brain has never been easy. With wires as thin as 10 nanometers being used, today’s conventional computer is reaching its ultimate limit. Therefore, inspired by the human brain, Professor Gimzewski attempted to develop a device resembling synapses and neurons. “The first step actually was to build such artificial synaptic connections using something called atomic switches,” states Professor Gimzewski. Incidentally, atomic switches were discovered by Professor Aono of MANA. It is a novel switching device, which works on the basis of movement of metal atoms/ions associated with their redox processes due to an applied potential.

The full article is available here.