Next Article in Journal
Ag-Nanowire Bundles with Gap Hot Spots Synthesized in Track-Etched Membranes as Effective SERS-Substrates
Next Article in Special Issue
Neuro-Inspired Computing with Spin-VCSELs
Previous Article in Journal
Self-Assembled MoS2/ssDNA Nanostructures for the Capacitive Aptasensing of Acetamiprid Insecticide
 
 
Article

Time-Multiplexed Spiking Convolutional Neural Network Based on VCSELs for Unsupervised Image Classification

1
Department of Information and Communication Systems Engineering, University of the Aegean, Palama 2, 83200 Samos, Greece
2
Department Informatics and Computer Engineering, University of West Attica, Ag. Spyridonos, 12243 Egaleo, Greece
*
Author to whom correspondence should be addressed.
Academic Editor: Ripalta Stabile
Appl. Sci. 2021, 11(4), 1383; https://doi.org/10.3390/app11041383
Received: 21 December 2020 / Revised: 23 January 2021 / Accepted: 29 January 2021 / Published: 3 February 2021
(This article belongs to the Special Issue Photonics for Optical Computing)
In this work, we present numerical results concerning a multilayer “deep” photonic spiking convolutional neural network, arranged so as to tackle a 2D image classification task. The spiking neurons used are typical two-section quantum-well vertical-cavity surface-emitting lasers that exhibit isomorphic behavior to biological neurons, such as integrate-and-fire excitability and timing encoding. The isomorphism of the proposed scheme to biological networks is extended by replicating the retina ganglion cell for contrast detection in the photonic domain and by utilizing unsupervised spike dependent plasticity as the main training technique. Finally, in this work we also investigate the possibility of exploiting the fast carrier dynamics of lasers so as to time-multiplex spatial information and reduce the number of physical neurons used in the convolutional layers by orders of magnitude. This last feature unlocks new possibilities, where neuron count and processing speed can be interchanged so as to meet the constraints of different applications. View Full-Text
Keywords: neuromorphic computing; optical neural networks; image classification; machine learning; laser dynamics; semiconductor lasers; VCSEL neuromorphic computing; optical neural networks; image classification; machine learning; laser dynamics; semiconductor lasers; VCSEL
Show Figures

Figure 1

MDPI and ACS Style

Skontranis, M.; Sarantoglou, G.; Deligiannidis, S.; Bogris, A.; Mesaritakis, C. Time-Multiplexed Spiking Convolutional Neural Network Based on VCSELs for Unsupervised Image Classification. Appl. Sci. 2021, 11, 1383. https://doi.org/10.3390/app11041383

AMA Style

Skontranis M, Sarantoglou G, Deligiannidis S, Bogris A, Mesaritakis C. Time-Multiplexed Spiking Convolutional Neural Network Based on VCSELs for Unsupervised Image Classification. Applied Sciences. 2021; 11(4):1383. https://doi.org/10.3390/app11041383

Chicago/Turabian Style

Skontranis, Menelaos, George Sarantoglou, Stavros Deligiannidis, Adonis Bogris, and Charis Mesaritakis. 2021. "Time-Multiplexed Spiking Convolutional Neural Network Based on VCSELs for Unsupervised Image Classification" Applied Sciences 11, no. 4: 1383. https://doi.org/10.3390/app11041383

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop