Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit

There is an ever-growing demand for artificial intelligence. Optical processors, which compute with photons instead of electrons, can fundamentally accelerate the development of artificial intelligence by offering substantially improved computing performance. There has been long-term interest in opt...

Full description

Saved in:
Bibliographic Details
Published in:Nature photonics Vol. 15; no. 5; pp. 367 - 373
Main Authors: Zhou Tiankuang, Lin, Xing, Wu, Jiamin, Chen Yitong, Xie Hao, Li, Yipeng, Fan Jingtao, Wu, Huaqiang, Lu, Fang, Dai Qionghai
Format: Journal Article
Language:English
Published: London Nature Publishing Group 01.05.2021
Subjects:
Adaptive systems
Artificial intelligence
Artificial neural networks
Complexity
Computation
Computer architecture
Data processing
High speed
Information processing
Light diffraction
Light speed
Linearity
Microprocessors
Neural networks
Object recognition
Optoelectronics
Photons
Reconfiguration
ISSN:1749-4885, 1749-4893
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:There is an ever-growing demand for artificial intelligence. Optical processors, which compute with photons instead of electrons, can fundamentally accelerate the development of artificial intelligence by offering substantially improved computing performance. There has been long-term interest in optically constructing the most widely used artificial-intelligence architecture, that is, artificial neural networks, to achieve brain-inspired information processing at the speed of light. However, owing to restrictions in design flexibility and the accumulation of system errors, existing processor architectures are not reconfigurable and have limited model complexity and experimental performance. Here, we propose the reconfigurable diffractive processing unit, an optoelectronic fused computing architecture based on the diffraction of light, which can support different neural networks and achieve a high model complexity with millions of neurons. Along with the developed adaptive training approach to circumvent system errors, we achieved excellent experimental accuracies for high-speed image and video recognition over benchmark datasets and a computing performance superior to that of cutting-edge electronic computing platforms.Linear diffractive structures are by themselves passive systems but researchers here exploit the non-linearity of a photodetector to realize a reconfigurable diffractive ‘processing’ unit. High-speed image and video recognition is demonstrated.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1749-4885
1749-4893
DOI:10.1038/s41566-021-00796-w