Neuromorphic photonics circuit for efficient multichannel signal coding
This study presents a model of a fully connected neural network (NN) implemented on elements of integrated photonics, including the software implementation of this architecture. The main purpose was to compare the results of two NNs, for this reason the average correlation coefficient was calculated, which amounted to 0.9904. This indicates a high degree of similarity and accuracy between the model's performance on PC and photonics elements.
Authors
- Ekaterina A. Lopukhova (ORCID: https://orcid.org/0000-0001-6802-6010)
- Ruslan V. Kutluyarov (ORCID: https://orcid.org/0000-0003-2724-6978)
- Aida G. Zakoyan (ORCID: https://orcid.org/0000-0002-6362-0755)
- Grigory S. Voronkov (ORCID: https://orcid.org/0000-0002-8788-2696)
- Ivan V. Stepanov (ORCID: https://orcid.org/0000-0003-4677-5808)
- Elizaveta P. Grakhova (ORCID: https://orcid.org/0000-0003-2207-2702)
Institutions
- Ufa Institute of Chemistry (RU)
Publication Details
- Published
- 2024-09-06
- DOI
- https://doi.org/10.1117/12.3026590
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00