Programmable chalcogenide-based all-optical deep neural networks
We demonstrate a passive all-chalcogenide all-optical perceptron scheme. The network’s nonlinear activation function (NLAF) relies on the nonlinear response of Ge2Sb2Te5 to femtosecond laser pulses. We measured the sub-picosecond time-resolved optical constants of Ge2Sb2Te5 at a wavelength of 1500 n...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2022-05-01
|
| Series: | Nanophotonics |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/nanoph-2022-0099 |
| _version_ | 1797789000293941248 |
|---|---|
| author | Teo Ting Yu Ma Xiaoxuan Pastor Ernest Wang Hao George Jonathan K. Yang Joel K. W. Wall Simon Miscuglio Mario Simpson Robert E. Sorger Volker J. |
| author_facet | Teo Ting Yu Ma Xiaoxuan Pastor Ernest Wang Hao George Jonathan K. Yang Joel K. W. Wall Simon Miscuglio Mario Simpson Robert E. Sorger Volker J. |
| author_sort | Teo Ting Yu |
| collection | DOAJ |
| description | We demonstrate a passive all-chalcogenide all-optical perceptron scheme. The network’s nonlinear activation function (NLAF) relies on the nonlinear response of Ge2Sb2Te5 to femtosecond laser pulses. We measured the sub-picosecond time-resolved optical constants of Ge2Sb2Te5 at a wavelength of 1500 nm and used them to design a high-speed Ge2Sb2Te5-tuned microring resonator all-optical NLAF. The NLAF had a sigmoidal response when subjected to different laser fluence excitation and had a dynamic range of −9.7 dB. The perceptron’s waveguide material was AlN because it allowed efficient heat dissipation during laser switching. A two-temperature analysis revealed that the operating speed of the NLAF is ≤1
$\le 1$
ns. The percepton’s nonvolatile weights were set using low-loss Sb2S3-tuned Mach Zehnder interferometers (MZIs). A three-layer deep neural network model was used to test the feasibility of the network scheme and a maximum training accuracy of 94.5% was obtained. We conclude that combining Sb2S3-programmed MZI weights with the nonlinear response of Ge2Sb2Te5 to femtosecond pulses is sufficient to perform energy-efficient all-optical neural classifications at rates greater than 1 GHz. |
| first_indexed | 2024-03-13T01:44:33Z |
| format | Article |
| id | doaj.art-431e55679a07411989ddfbb1bea684f6 |
| institution | Directory Open Access Journal |
| issn | 2192-8614 |
| language | English |
| last_indexed | 2024-03-13T01:44:33Z |
| publishDate | 2022-05-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Nanophotonics |
| spelling | doaj.art-431e55679a07411989ddfbb1bea684f62023-07-03T10:20:07ZengDe GruyterNanophotonics2192-86142022-05-0111174073408810.1515/nanoph-2022-0099Programmable chalcogenide-based all-optical deep neural networksTeo Ting Yu0Ma Xiaoxuan1Pastor Ernest2Wang Hao3George Jonathan K.4Yang Joel K. W.5Wall Simon6Miscuglio Mario7Simpson Robert E.8Sorger Volker J.9Singapore University of Technology and Design, 8 Somapah Road, Singapore487372, SingaporeDeptartment of Electrical and Computer Engineering, George Washington University, Washington, DC, USAICFO - Institut de Ciencies Fotoniques, The Barcelona Institute of Science and Technology, Av. Carl Friedrich Gauss 3, Castelldefels08860, Barcelona, SpainSingapore University of Technology and Design, 8 Somapah Road, Singapore487372, SingaporeDeptartment of Electrical and Computer Engineering, George Washington University, Washington, DC, USASingapore University of Technology and Design, 8 Somapah Road, Singapore487372, SingaporeICFO - Institut de Ciencies Fotoniques, The Barcelona Institute of Science and Technology, Av. Carl Friedrich Gauss 3, Castelldefels08860, Barcelona, SpainDeptartment of Electrical and Computer Engineering, George Washington University, Washington, DC, USASingapore University of Technology and Design, 8 Somapah Road, Singapore487372, SingaporeDeptartment of Electrical and Computer Engineering, George Washington University, Washington, DC, USAWe demonstrate a passive all-chalcogenide all-optical perceptron scheme. The network’s nonlinear activation function (NLAF) relies on the nonlinear response of Ge2Sb2Te5 to femtosecond laser pulses. We measured the sub-picosecond time-resolved optical constants of Ge2Sb2Te5 at a wavelength of 1500 nm and used them to design a high-speed Ge2Sb2Te5-tuned microring resonator all-optical NLAF. The NLAF had a sigmoidal response when subjected to different laser fluence excitation and had a dynamic range of −9.7 dB. The perceptron’s waveguide material was AlN because it allowed efficient heat dissipation during laser switching. A two-temperature analysis revealed that the operating speed of the NLAF is ≤1 $\le 1$ ns. The percepton’s nonvolatile weights were set using low-loss Sb2S3-tuned Mach Zehnder interferometers (MZIs). A three-layer deep neural network model was used to test the feasibility of the network scheme and a maximum training accuracy of 94.5% was obtained. We conclude that combining Sb2S3-programmed MZI weights with the nonlinear response of Ge2Sb2Te5 to femtosecond pulses is sufficient to perform energy-efficient all-optical neural classifications at rates greater than 1 GHz.https://doi.org/10.1515/nanoph-2022-0099all-optical deep neural networkchalcogenide reconfigurable photonicsultra-fast dynamic response of phase change material |
| spellingShingle | Teo Ting Yu Ma Xiaoxuan Pastor Ernest Wang Hao George Jonathan K. Yang Joel K. W. Wall Simon Miscuglio Mario Simpson Robert E. Sorger Volker J. Programmable chalcogenide-based all-optical deep neural networks Nanophotonics all-optical deep neural network chalcogenide reconfigurable photonics ultra-fast dynamic response of phase change material |
| title | Programmable chalcogenide-based all-optical deep neural networks |
| title_full | Programmable chalcogenide-based all-optical deep neural networks |
| title_fullStr | Programmable chalcogenide-based all-optical deep neural networks |
| title_full_unstemmed | Programmable chalcogenide-based all-optical deep neural networks |
| title_short | Programmable chalcogenide-based all-optical deep neural networks |
| title_sort | programmable chalcogenide based all optical deep neural networks |
| topic | all-optical deep neural network chalcogenide reconfigurable photonics ultra-fast dynamic response of phase change material |
| url | https://doi.org/10.1515/nanoph-2022-0099 |
| work_keys_str_mv | AT teotingyu programmablechalcogenidebasedallopticaldeepneuralnetworks AT maxiaoxuan programmablechalcogenidebasedallopticaldeepneuralnetworks AT pastorernest programmablechalcogenidebasedallopticaldeepneuralnetworks AT wanghao programmablechalcogenidebasedallopticaldeepneuralnetworks AT georgejonathank programmablechalcogenidebasedallopticaldeepneuralnetworks AT yangjoelkw programmablechalcogenidebasedallopticaldeepneuralnetworks AT wallsimon programmablechalcogenidebasedallopticaldeepneuralnetworks AT miscugliomario programmablechalcogenidebasedallopticaldeepneuralnetworks AT simpsonroberte programmablechalcogenidebasedallopticaldeepneuralnetworks AT sorgervolkerj programmablechalcogenidebasedallopticaldeepneuralnetworks |