Presentation
20 June 2024 Photonic spiking neurons and spiking neural networks
Dafydd Owen-Newns, Andrew Adair, Dylan Black, Giovanni Donati, Joshua Robertson, Antonio Hurtado
Author Affiliations +
Abstract
Photonic approaches emulating the powerful computational capabilities of the brain are receiving increasing research interest for radically new paradigms in ultrafast neuromorphic (brain-like) information processing and Artificial Intelligence (AI). This talk will report our research on light-enabled neuromorphic systems built with artificial photonic spiking neurons and photonic spiking neural networks (SNN). We will review the properties and performance of the photonic devices employed for the implementation of optical spiking neurons, including semiconductor lasers (e.g. Vertical Cavity Surface Emitting Lasers) and resonant tunnelling diodes. We will also discuss the strategies for their network-connectivity into photonic SNN architectures, and the techniques and algorithms realised for their use in complex functional information processing tasks (e.g. pattern recognition, image processing, data classification). We will also showcase the potentials of these spike-based photonic processing systems for ultrafast, low-energy and high-accuracy performance, with a hardware-friendly implementation that benefits from spike-based learning protocols with highly-reduced complexity.
Conference Presentation
© (2024) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Dafydd Owen-Newns, Andrew Adair, Dylan Black, Giovanni Donati, Joshua Robertson, and Antonio Hurtado "Photonic spiking neurons and spiking neural networks", Proc. SPIE PC13002, Semiconductor Lasers and Laser Dynamics XI, PC1300201 (20 June 2024); https://doi.org/10.1117/12.3025405
Advertisement
Advertisement
KEYWORDS
Neurons

Artificial neural networks

Data processing

Artificial intelligence

Ultrafast phenomena

Network architectures

Pattern recognition

RELATED CONTENT


Back to Top