All Optical Neural Networks for Low Power Edge Computing
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
We developed a simplistic physics-based model of an all-optical neural network that mimics the encoder part of an autoencoder neural network for image compression. Our approach relies on the generation of a MATLAB-based model for both data compression and decompression and utilizes MATLAB's built-in autoencoder networks in combination with simple propagation of optical fields between layers constituting phase elements via Fourier transform. We optimize the phase elements using the particle swarm optimization technique and using our model, we demonstrate a compression ratio of 25% for 2828-pixel input images containing numeric digits from 0 to 9.
- Research Organization:
- Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE National Nuclear Security Administration (NNSA)
- DOE Contract Number:
- NA0003525
- OSTI ID:
- 1832286
- Report Number(s):
- SAND2021-13776; 701747
- Country of Publication:
- United States
- Language:
- English
Similar Records
Complexity optimized vector quantization: A neural network approach
Complexity optimized vector quantization: A neural network approach
An analysis of image storage systems for scalable training of deep neural networks
Conference
·
Thu Jan 30 23:00:00 EST 1992
·
OSTI ID:10189003
Complexity optimized vector quantization: A neural network approach
Conference
·
Thu Jan 30 23:00:00 EST 1992
·
OSTI ID:7173635
An analysis of image storage systems for scalable training of deep neural networks
Conference
·
Thu Dec 31 23:00:00 EST 2015
·
OSTI ID:1335300