comment on this article

The missing piece of the puzzle for artificial brains?

Credit: NIST

A superconducting switch that learns like a biological system has been developed by researchers at the National Institute of Standards and Technology (NIST).

NIST claims that its switch has the potential to operate like the human brain, connecting processors and storing memories in future computers.

The switch is called a synapse, named after its biological counterpart and the team claims it supplies a missing piece for so-called neuromorphic computers. The hope is that this artificial synapse could improve perception and decision-making for applications such as autonomous vehicles.

This synapse has the ability process incoming electrical spikes to customise spiking output signals, says NIST. This processing is based on a flexible internal design that can be tuned by experience or its environment. The more firing between cells or processors, the stronger the connection. According to NIST, this means the synapse can maintain old circuits, as well as creating new ones.

The NIST synapse is said to be able to fire 1 billion times per second – compared to a brain cell's 50 times per second – using just a whiff of energy, about one ten-thousandth as much as a human synapse.

The spiking energy is less than 1 attojoule, lower than the background energy at room temperature and on a par with the chemical energy bonding two atoms in a molecule.

"The NIST synapse has lower energy needs than the human synapse, and we don't know of any other artificial synapse that uses less energy," NIST physicist, Mike Schneider says.

The goal is to use this synapse in neuromorphic computers made of superconducting components, which can transmit electricity without resistance. Data would be transmitted, processed and stored in units of magnetic flux. NIST believes this would be more efficient than other designs based on semiconductors or software.

Although superconducting devices mimicking brain cells and transmission lines have been developed, the team claims that efficient synapses have been missing.

A computer processes data in sequence and stores memory in a separate unit, whereas the brain processes data both in sequence and simultaneously and stores memories in synapses all over the system.

The NIST synapse is a Josephson junction. These junctions are a sandwich of superconducting materials with an insulator as a filling. Voltage spikes are produced when electrical current through the junction exceeds the critical current. The synapse uses standard niobium electrodes, but has a filling made of nanoscale clusters of manganese in a silicon matrix.

The nanoclusters act like tiny bar magnets with spins that can be oriented either randomly or in a coordinated manner.

"These are customised Josephson junctions," Schneider explains. "We can control the number of nanoclusters pointing in the same direction, which affects the superconducting properties of the junction."

The synapse rests in a superconducting state, except when it's activated by incoming current.

Researchers apply current pulses in a magnetic field, so a number of nanoclusters point in the same direction – in other words, increase the magnetic ordering. This magnetic effect is said to gradually reduce the critical current level, making it easier to create a normal conductor and produce voltage spikes.

Pulses can also be applied without a magnetic field to reduce the magnetic ordering and raise the critical current. This design, in which different inputs alter the spin alignment and resulting output signals, is apparently similar to how the brain operates.

Synapse behaviour can also be tuned by changing how the device is made and its operating temperature. NIST says it can reduce the pulse energy needed to raise or lower the magnetic order of the device by making the nanoclusters smaller.

The synapses can also be stacked in 3D to make large systems, which the team believe could be used for computing. The researchers say they have created a circuit model to simulate how such a system would operate.

The NIST team hope that its synapse could pave the way for a more complex neuromorphic system, than the ones which have already been demonstrated with other technologies.

Author
Bethan Grylls

Comment on this article


Websites

https://www.nist.gov

Companies

NIST

This material is protected by MA Business copyright See Terms and Conditions. One-off usage is permitted but bulk copying is not. For multiple copies contact the sales team.

What you think about this article:


Add your comments

Name
 
Email
 
Comments
 

Your comments/feedback may be edited prior to publishing. Not all entries will be published.
Please view our Terms and Conditions before leaving a comment.

Related Articles