preprints.org > physical sciences > optics and photonics > doi: 10.20944/preprints202410.1758.v1

Preprint Review Version 1 This version is not peer-reviewed

Version 1 : Received: 21 October 2024 / Approved: 22 October 2024 / Online: 24 October 2024 (12:11:01 CEST)

Drawing inspiration from biological brain's energy-efficient information-processing mechanisms, photonic integrated circuits (PIC) have facilitated the development of ultrafast artificial neural networks. This in turn is envisaged to offer potential solutions to the growing demand for artificial intelligence employing machine learning in various domains, from nonlinear optimization and telecommunication to medical diagnosis. At the meantime, silicon photonics has emerged as a mainstream technology for integrated chip based application. However, challenges still need to be addressed in scaling it further for broader applications due to the requirement of co-integration of electronic circuitry for control and calibration. Leveraging physics in algorithms and nanoscale materials holds promise for achieving low-power, miniaturized chips capable of real-time inference and learning. In this back drop, we present the state of the art in neuromorphic photonic computing, focusing primarily on architecture, weighting mechanisms, photonic neurons, and training while giving an over-all view on recent advancements, challenges, and prospects. We also emphasize and high light the need for revolutionary hardware innovations to scale up neuromorphic systems while enhancing energy efficiency and performance.

Neuromorphic Photonics; Photonic Integrated Circuits; Optical Computing; Silicon Photonics; On-Chip Machine Learning; Neuromorphic Computing; Photonic Neural Networks; Integrated circuits

Physical Sciences, Optics and Photonics

* All users must log in before leaving a comment