Presentation
18 June 2024 Coherent multi-frequency photonic activation function
Grigorii Slinkov, Steven Becker, Dirk Englund, Birgit Stiller
Author Affiliations +
Abstract
The advent of photonic neural networks posed a challenge of designing an activation function that would unlock the full potential of photonics (low-latency, energy-efficiency, WDM) for machine learning applications. In this work we demonstrate a coherent, multi-frequency photonic nonlinear activation function based on stimulated Brillouin scattering . These properties not only make it compatible with existing MZI mesh-based phase-reliant optical matrix multiplication schemes, but also facilitate resource-efficient frequency-basis information encoding. Our design features all-optical activation function shape tuning and is capable of providing net gain, compensating for insertion losses.
Conference Presentation
© (2024) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Grigorii Slinkov, Steven Becker, Dirk Englund, and Birgit Stiller "Coherent multi-frequency photonic activation function", Proc. SPIE PC13017, Machine Learning in Photonics, PC130170J (18 June 2024); https://doi.org/10.1117/12.3017756
Advertisement
Advertisement
KEYWORDS
Data modeling

Neural networks

Machine learning

Matrix multiplication

Optoelectronics

Performance modeling

Power consumption

Back to Top