Home / Blog / Atomic switch networks

Atomic switch networks

Computer are faster, but brains are smarter. And don’t think about our brain. All brains are smarter than computers: a fly can take decision on the best fly route and avoidance with just 5,000 neurones and do that using an infinitesimal fraction of the power needed to a flight management system on a plane to take equivalent decisions (don’t let yourself be fooled by the size of a fly versus the size of an airplane. For this type of decision the size is irrelevant).
The reasons lies in the "hardware" the way information is being processed. In a computer there is a basically sequential but accurate processing of data stored in very specific cells, such that 1+1=2. In a brain the processing and storage cannot be separated, everything occurs in parallel (to the point that where sequentialisation is mandatory special neuronal circuits have to be involved) and for a brain 1+1 is probably equal 2 but…
A point in case is the text in the figure. If you just read it … you read it. Only when you start looking at individual words and letters you realise it is a complete mess. The brain is NOT taking care of rearranging the letters, it simply skip the letter order since it reads in parallel and readjust them in the same way it always has to rearrange what it is being read.
A joint team of researchers from the California NanoSystems Institute, UCLA and International Center for Materials Nanoarchitectures (MANA) at the National Institute for Material Science in Japan have developed a self assembled neuromorphic device composed by over a billion of interconnected atomic-switch inorganic synapses (ASN). The interconnection is based on a complex networks of silver nanowires.
The ASN structure is such that, like in our brain, processing and storage are a single thing and the whole network shows a emergent behaviour as result of its “working” (not of its processing as in a computer where you get a “result” at the end; here the “result” is a changed state of the ASN itself).
These kind of devices have significant potential to be used as for neuromorphic computing to support cognitive processes.
Only few days ago I posted a few thoughts on the possibility that a computer might eventually “feel”  as we do saying that their computational architectures do not support the emergence of feeling but that other types of architectures might. Well this might the case for ASN.

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the New Initiative Committee and co-chairs the Digital Reality Initiative. He is a member of the IEEE in 2050 Ad Hoc Committee. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.