Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors
Diffractive Neural Networks (DNNs) leverage the power of light to enhance computational performance in machine learning, offering a pathway to high-speed, low-energy, and large-scale neural information processing. However, most existing DNN architectures are optimized for single tasks and thus lack...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2024-10-01
|
| Series: | Nanophotonics |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/nanoph-2024-0483 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823860479729074176 |
|---|---|
| author | Behroozinia Sahar Gu Qing |
| author_facet | Behroozinia Sahar Gu Qing |
| author_sort | Behroozinia Sahar |
| collection | DOAJ |
| description | Diffractive Neural Networks (DNNs) leverage the power of light to enhance computational performance in machine learning, offering a pathway to high-speed, low-energy, and large-scale neural information processing. However, most existing DNN architectures are optimized for single tasks and thus lack the flexibility required for the simultaneous execution of multiple tasks within a unified artificial intelligence platform. In this work, we utilize the polarization and wavelength degrees of freedom of light to achieve optical multi-task identification using the MNIST, FMNIST, and KMNIST datasets. Employing bilayer cascaded metasurfaces, we construct dual-channel DNNs capable of simultaneously classifying two tasks, using polarization and wavelength multiplexing schemes through a meta-atom library. Numerical evaluations demonstrate performance accuracies comparable to those of individually trained single-channel, single-task DNNs. Extending this approach to three-task parallel recognition reveals an expected performance decline yet maintains satisfactory classification accuracies of greater than 80 % for all tasks. We further introduce a novel end-to-end joint optimization framework to redesign the three-task classifier, demonstrating substantial improvements over the meta-atom library design and offering the potential for future multi-channel DNN designs. Our study could pave the way for the development of ultrathin, high-speed, and high-throughput optical neural computing systems. |
| format | Article |
| id | doaj-art-2194deef65464c3b99b028b02d9cda80 |
| institution | Kabale University |
| issn | 2192-8614 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Nanophotonics |
| spelling | doaj-art-2194deef65464c3b99b028b02d9cda802025-02-10T13:24:47ZengDe GruyterNanophotonics2192-86142024-10-0113244505451710.1515/nanoph-2024-0483Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processorsBehroozinia Sahar0Gu Qing1Department of Electrical and Computer Engineering, North Carolina State University, Raleigh, 27695, USADepartment of Electrical and Computer Engineering, North Carolina State University, Raleigh, 27695, USADiffractive Neural Networks (DNNs) leverage the power of light to enhance computational performance in machine learning, offering a pathway to high-speed, low-energy, and large-scale neural information processing. However, most existing DNN architectures are optimized for single tasks and thus lack the flexibility required for the simultaneous execution of multiple tasks within a unified artificial intelligence platform. In this work, we utilize the polarization and wavelength degrees of freedom of light to achieve optical multi-task identification using the MNIST, FMNIST, and KMNIST datasets. Employing bilayer cascaded metasurfaces, we construct dual-channel DNNs capable of simultaneously classifying two tasks, using polarization and wavelength multiplexing schemes through a meta-atom library. Numerical evaluations demonstrate performance accuracies comparable to those of individually trained single-channel, single-task DNNs. Extending this approach to three-task parallel recognition reveals an expected performance decline yet maintains satisfactory classification accuracies of greater than 80 % for all tasks. We further introduce a novel end-to-end joint optimization framework to redesign the three-task classifier, demonstrating substantial improvements over the meta-atom library design and offering the potential for future multi-channel DNN designs. Our study could pave the way for the development of ultrathin, high-speed, and high-throughput optical neural computing systems.https://doi.org/10.1515/nanoph-2024-0483metasurfacediffractive neural networkdeep learning |
| spellingShingle | Behroozinia Sahar Gu Qing Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors Nanophotonics metasurface diffractive neural network deep learning |
| title | Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors |
| title_full | Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors |
| title_fullStr | Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors |
| title_full_unstemmed | Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors |
| title_short | Leveraging multiplexed metasurfaces for multi-task learning with all-optical diffractive processors |
| title_sort | leveraging multiplexed metasurfaces for multi task learning with all optical diffractive processors |
| topic | metasurface diffractive neural network deep learning |
| url | https://doi.org/10.1515/nanoph-2024-0483 |
| work_keys_str_mv | AT behrooziniasahar leveragingmultiplexedmetasurfacesformultitasklearningwithallopticaldiffractiveprocessors AT guqing leveragingmultiplexedmetasurfacesformultitasklearningwithallopticaldiffractiveprocessors |