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

• 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, Fang, Lu, Dai, Qionghai
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.05.2021; Nature Publishing Group
• Subjects: 639/624/1075; 639/624/1107/510; Adaptive systems; Applied and Technical Physics; 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; Physics; Physics and Astronomy; Quantum Physics; Reconfiguration
• ISSN: 1749-4885; 1749-4893
• DOI: 10.1038/s41566-021-00796-w

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.