Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh
Abstract Photonic accelerators have risen as energy efficient, low latency counterparts to computational hungry digital modules for machine learning applications. On the other hand, upscaling integrated photonic circuits to meet the demands of state-of-the-art machine learning schemes such as convol...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-04-01
|
| Series: | Communications Engineering |
| Online Access: | https://doi.org/10.1038/s44172-025-00416-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849234527397871616 |
|---|---|
| author | Aris Tsirigotis George Sarantoglou Stavros Deligiannidis Erica Sánchez David Sanchez Ana Gutierrez Adonis Bogris Jose Capmany Charis Mesaritakis |
| author_facet | Aris Tsirigotis George Sarantoglou Stavros Deligiannidis Erica Sánchez David Sanchez Ana Gutierrez Adonis Bogris Jose Capmany Charis Mesaritakis |
| author_sort | Aris Tsirigotis |
| collection | DOAJ |
| description | Abstract Photonic accelerators have risen as energy efficient, low latency counterparts to computational hungry digital modules for machine learning applications. On the other hand, upscaling integrated photonic circuits to meet the demands of state-of-the-art machine learning schemes such as convolutional layers, remains challenging. In this work, we experimentally validate a photonic-integrated neuromorphic accelerator that uses a hardware-friendly optical spectrum slicing technique through a reconfigurable silicon photonic mesh. The proposed scheme acts as an analogue convolutional engine, enabling information preprocessing in the optical domain, dimensionality reduction, and extraction of spatio-temporal features. Numerical results demonstrate that with only 7 photonic nodes, critical modules of a digital convolutional neural network can be replaced. As a result, a 98.6% accuracy on the MNIST dataset was numerically achieved, with an estimation of power consumption reduction up to 30% compared to digital convolutional neural networks. Experimental results using a reconfigurable silicon integrated chip confirm these findings, achieving 97.7% accuracy with only three optical nodes. |
| format | Article |
| id | doaj-art-1d45e67dbafe488aa855b2d9f54bf449 |
| institution | Kabale University |
| issn | 2731-3395 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Communications Engineering |
| spelling | doaj-art-1d45e67dbafe488aa855b2d9f54bf4492025-08-20T04:03:07ZengNature PortfolioCommunications Engineering2731-33952025-04-014111110.1038/s44172-025-00416-3Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable meshAris Tsirigotis0George Sarantoglou1Stavros Deligiannidis2Erica Sánchez3David Sanchez4Ana Gutierrez5Adonis Bogris6Jose Capmany7Charis Mesaritakis8Department of Information and Communication Systems Engineering, University of the AegeanDepartment of Information and Communication Systems Engineering, University of the AegeanDepartment of Informatics and Computer Engineering, University of West AtticaiPronics Programmable PhotonicsiPronics Programmable PhotonicsiPronics Programmable PhotonicsDepartment of Informatics and Computer Engineering, University of West AtticaPhotonics Research Labs, Universitat Politècnica de ValènciaDepartment of Biomedical Engineering, University of West AtticaAbstract Photonic accelerators have risen as energy efficient, low latency counterparts to computational hungry digital modules for machine learning applications. On the other hand, upscaling integrated photonic circuits to meet the demands of state-of-the-art machine learning schemes such as convolutional layers, remains challenging. In this work, we experimentally validate a photonic-integrated neuromorphic accelerator that uses a hardware-friendly optical spectrum slicing technique through a reconfigurable silicon photonic mesh. The proposed scheme acts as an analogue convolutional engine, enabling information preprocessing in the optical domain, dimensionality reduction, and extraction of spatio-temporal features. Numerical results demonstrate that with only 7 photonic nodes, critical modules of a digital convolutional neural network can be replaced. As a result, a 98.6% accuracy on the MNIST dataset was numerically achieved, with an estimation of power consumption reduction up to 30% compared to digital convolutional neural networks. Experimental results using a reconfigurable silicon integrated chip confirm these findings, achieving 97.7% accuracy with only three optical nodes.https://doi.org/10.1038/s44172-025-00416-3 |
| spellingShingle | Aris Tsirigotis George Sarantoglou Stavros Deligiannidis Erica Sánchez David Sanchez Ana Gutierrez Adonis Bogris Jose Capmany Charis Mesaritakis Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh Communications Engineering |
| title | Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh |
| title_full | Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh |
| title_fullStr | Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh |
| title_full_unstemmed | Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh |
| title_short | Photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh |
| title_sort | photonic neuromorphic accelerator for convolutional neural networks based on an integrated reconfigurable mesh |
| url | https://doi.org/10.1038/s44172-025-00416-3 |
| work_keys_str_mv | AT aristsirigotis photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT georgesarantoglou photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT stavrosdeligiannidis photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT ericasanchez photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT davidsanchez photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT anagutierrez photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT adonisbogris photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT josecapmany photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh AT charismesaritakis photonicneuromorphicacceleratorforconvolutionalneuralnetworksbasedonanintegratedreconfigurablemesh |