Neuromorphic photonic technologies and architectures: scaling opportunities and performance frontiers [Invited]
- 1 June 2022
- journal article
- review article
- Published by Optica Publishing Group in Optical Materials Express
- Vol. 12 (6) , 2343-2367
Abstract
We review different technologies and architectures for neuromorphic photonic accelerators, spanning from bulk optics to photonic-integrated-circuits (PICs), and assess compute efficiency in OPs/Watt through the lens of a comparative study where key technology aspects are analyzed. With an emphasis on PIC neuromorphic accelerators, we shed light onto the latest advances in photonic and plasmonic modulation technologies for the realization of weighting elements in training and inference applications, and present a recently introduced scalable coherent crossbar layout. Finally, we stress that current technologies face challenges endowing photonic accelerators with compute efficiencies in the PetaOPs/W, and discuss future implementation pathways towards improving performance.Keywords
Funding Information
- Hellenic Foundation for Research and Innovation (DeepLight (4233))
- H2020 LEIT Information and Communication Technologies (PLASMONIAC (871391))
This publication has 87 references indexed in Scilit:
- High-speed plasmonic phase modulatorsNature Photonics, 2014
- Thermo-optic control of dielectric-loaded plasmonic Mach–Zehnder interferometers and directional coupler switchesNanotechnology, 2012
- Active plasmonics in WDM traffic switching applicationsScientific Reports, 2012
- Efficient thermo-optically controlled Mach-Zhender interferometers using dielectric-loaded plasmonic waveguidesOptics Express, 2012
- The Case for PlasmonicsScience, 2010
- Plasmonics beyond the diffraction limitNature Photonics, 2010
- Plasmonics: the next chip-scale technologyMaterials Today, 2006
- Experimental realization of any discrete unitary operatorPhysical Review Letters, 1994
- Quantum-confined Stark effect in InGaAs/InP quantum wells grown by organometallic vapor phase epitaxyApplied Physics Letters, 1987
- A logical calculus of the ideas immanent in nervous activityBulletin of Mathematical Biology, 1943