Home / Blog / Artificial synapse, better than the natural one, but not so good …

Artificial synapse, better than the natural one, but not so good …

This artificial synapse designed for neuromorphic computing mimics the operation of switch between two neurons. One artificial synapse is located at the center of each X. This chip is 1 square centimeter in size. (The thick black vertical lines are electrical probes used for testing.) Credit: NIST

Researchers at NIST have built an artificial synapses mimicking the ones operating in the brain (ours and the ones of any other animal): it can learn (i.e. its responses to stimuli depends on previous stimuli) like our synapses but it has not been created to become a component in a future (very future) artificial brain, rather to improve neuromorphic  computers.

In a way it is “better” than our synapse: it can take a decision (switch from one state to another) in billionth of a second, versus our “slow” ones that take twenty times as much, 50 ms. It is also more energy savvy, using just 1 ten-thousandth of the energy required by a human synapse to generate a spike.

On the other hand it is a sort of metallic cylinder 10 µm in diameter, that is 10 times bigger than our synapse.
Per sé, synapses are not particularly smart, nor a neurone in really smart (our neurones are as smart as a snail neurones…): the smartness is a consequence of their sheer number and interconnection. Being able to pack a huge amount of synapses, neurones and tightly connecting them is what makes the difference. To give an idea of the complexity using a different measuring stick consider that to sort out all the interconnections among the 300 million synapses contained in one cubic mm of our brain you would need to take very this sheets about 30 nanometer thick each one requiring some 10-20GB to store the connectivity information, meaning that to store the connections with one cubic mm of brain we would need about a PetaByte of data!

The sheer complexity of natural neural circuits are still beyond our technological capability and a few observers commenting on the result achieved by this NIST endeavour point out the difficulty of scaling up, moving from a few thousands synapses to several billions. True these artificial synapses are way faster than natural ones but being fast is not the crucial factors and it does not make for  the lower complexity.

Complex systems are inherently different form other systems, it is not about being bigger or faster, it is a qualitative difference. It is what creates the emergence of intelligence, perception and … sense of self, self awareness.

We are still far from that and we are still far from a clear understanding that pinpoints when a system is sufficiently complex to change its quality.

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.