Neuromorphic computing aims to mimic the architecture of the human brain to carry out computational tasks that are challenging and much more energy consuming for standard hardware. Despite progress in several fields of physics and engineering, the realization of artificial neural networks, which combine high operating speeds with fast and low-energy adaptability, remains a challenge. Here, we demonstrate an opto-magnetic neural network capable of learning and classification of digitized 3 × 3 characters exploiting local storage in the magnetic material. Using picosecond laser pulses, we find that micrometer sized synapses absorb well below 100 picojoule per synapse per laser pulse, with favorable scaling to smaller spatial dimensions. We, thus, managed to combine the speed and low-dissipation of optical networks with the low-energy adaptability and non-volatility of magnetism, providing a promising approach to fast and energy-efficient neuromorphic computing.
Skip Nav Destination
Article navigation
10 January 2022
Research Article|
January 10 2022
Training and pattern recognition by an opto-magnetic neural network
Special Collection:
Neuromorphic Computing: From Quantum Materials to Emergent Connectivity
A. Chakravarty;
A. Chakravarty
1
Radboud University, Institute for Molecules and Materials
, Nijmegen, The Netherlands
2
Charles University, Faculty of Mathematics and Physics
, Prague, The Czech Republic
Search for other works by this author on:
J. H. Mentink
;
J. H. Mentink
a)
1
Radboud University, Institute for Molecules and Materials
, Nijmegen, The Netherlands
a)Author to whom correspondence should be addressed: j.mentink@science.ru.nl
Search for other works by this author on:
S. Semin;
S. Semin
1
Radboud University, Institute for Molecules and Materials
, Nijmegen, The Netherlands
Search for other works by this author on:
Th. Rasing
Th. Rasing
1
Radboud University, Institute for Molecules and Materials
, Nijmegen, The Netherlands
Search for other works by this author on:
a)Author to whom correspondence should be addressed: j.mentink@science.ru.nl
Note: This paper is part of the APL Special Collection on Neuromorphic Computing: From Quantum Materials to Emergent Connectivity.
Appl. Phys. Lett. 120, 022403 (2022)
Article history
Received:
September 29 2021
Accepted:
December 22 2021
Citation
A. Chakravarty, J. H. Mentink, S. Semin, Th. Rasing; Training and pattern recognition by an opto-magnetic neural network. Appl. Phys. Lett. 10 January 2022; 120 (2): 022403. https://doi.org/10.1063/5.0073280
Download citation file:
Sign in
Don't already have an account? Register
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
Pay-Per-View Access
$40.00
Citing articles via
Roadmap on photonic metasurfaces
Sebastian A. Schulz, Rupert. F. Oulton, et al.
Feedback cooling of an insulating high-Q diamagnetically levitated plate
S. Tian, K. Jadeja, et al.
Special topic on Wide- and ultrawide-bandgap electronic semiconductor devices
Joachim Würfl, Tomás Palacios, et al.
Related Content
Supervised learning of an opto-magnetic neural network with ultrashort laser pulses
Appl. Phys. Lett. (May 2019)
Low-power, multiport, ultrafast, parametric switching in cascaded waveguide couplers
Appl. Phys. Lett. (June 2005)
Organic non‐linear optics and opto‐electronics
AIP Conference Proceedings (December 2010)
An opto-mechanical coupled-ring reflector driven by optical force for lasing wavelength control
Appl. Phys. Lett. (February 2016)
Ultrafast opto-magnetic effects induced by nitrogen-vacancy centers in diamond crystals
APL Photonics (June 2022)