Phase input of information in a 4F system with a diffraction neural network

Skidanov, R.V., Khanenko Y.V., Morozov A.E., Pronin A.S., Sorokin D.M., Doskolovich, L.L.

For Russian citation (Opticheskii Zhurnal):

Скиданов Р.В., Ханенко Ю.В., Морозов А.Е., Пронин А.С., Сорокин Д.М., Досколович Л.Л. Фазовый ввод информации в системе 4F с дифракционной нейронной сетью // Оптический журнал. 2026. Т. 93. № 5. С. 82–89. http://doi.org/10.17586/1023-5086-2026-93-05-82-89

Skidanov R.V., Khanenko Yu.V., Morozov A.E., Pronin A.S., Sorokin D.M., Doskolovich L.L. Phase input of information in a 4F system with a diffraction neural network [in Russian] // Opticheskii Zhurnal. 2026. V. 93. № 5. P. 82–89. http://doi.org/10.17586/1023-5086-2026-93-05-82-89

For citation (Journal of Optical Technology):

Abstract:

The subject of the study. Entering an object image into a 4F system for subsequent classification using a diffraction neural network. The purpose of the work. Confirming the possibility of effectively entering information into a 4F electromagnetic wave phase modulation system. Method. The optical scheme and energy efficiency of the system used are considered. Both mathematical modeling of the 4F system and a full-scale experiment in which the 4F system was assembled as a compact device were used. The main results. It is shown that the modulation of the electromagnetic wave phase at the input to the 4F information input system gives almost identical results in terms of classification reliability with significantly higher energy efficiency. Practical significance. The results obtained make it possible to significantly simplify the optical scheme for implementing diffraction neural networks in a 4F system and significantly increase the energy efficiency of computing.

Keywords:

4F system, diffraction neural network, object classification, spatial light modulator, phase modulation of an electromagnetic wave

Acknowledgements:

the research was carried out within the framework of the scientific program of the National Center for Physics and Mathematics, direction № 1 “National Center for Research on Supercomputer Architectures” Stage 2023-2025

OCIS codes: 070.0070, 050.0050

References:

1. Lin X., Rivenson Y., Yardimci N.T., et al. All-optical machine learning using diffractive deep neural networks // Science. 2018. V. 361. P. 1004–1008. https://doi.org/10.1126/science.aat8084 2. Goi E., Schoenhardt S., Gu M. Direct retrieval of Zernike-based pupil functions using integrated diffractive deep neural networks // Nat. Commun. 2022. V. 13. P. 7531. https://doi.org/10.1038/s41467-022-35349-4 3. Goi E., Chen X., Zhang Q.M., et al. Nanoprinted high-neuron-density optical linear perceptrons performing near-infrared inference on a CMOS chip // Light Sci. Appl. 2021. V. 10. P. 40. https://doi.org/10.1038/s41377-021-00483-z 4. Chen R.Y., Li Y.J., Lou M.H., et al. Physics-aware machine learning and adversarial attack in complex-valued reconfigurable diffractive all-optical neural network // Laser Photonics Rev. 2022. V. 16. P. 2200348. https://doi.org/10.48550/arXiv.2203.06055 5. Georgi P., Wei Q.S., Sain B., et al. Optical secret sharing with cascaded metasurface holography // Sci. Adv. 2021. V. 7. P. eabf9718. https://doi.org/10.1126/sciadv.abf9718

6. Zhang Y.C., Zhang Q.M., Yu H.Y., et al. Memory-less scattering imaging with ultrafast convolutional optical neural networks // Sci. Adv. 2024. V. 10. P. eadn2205. https://doi.org/10.1126/sciadv.adn2205

7. Li J.X., Mengu D., Yardimci N.T., et al. Spectrally encoded single-pixel machine vision using diffractive networks // Sci. Adv. 2021. V. 7. P. eabd7690. https://doi.org/10.1126/sciadv.abd7690

8. Mengu D., Tabassum A., Jarrahi M., et al. Snapshot multispectral imaging using a diffractive optical network // Light Sci. Appl. 2023. V. 12. P. 86. https://doi.org/10.1038/s41377-023-01135-0

9. Luo Y., Zhao Y.F., Li J.X., et al. Computational imaging without a computer: Seeing through random diffusers at the speed of light // Elight. 2022. V. 2. P. 4. https://doi.org/10.48550/arXiv.2107.06586

10. Lin X., Rivenson Y., Yardimci N.T., et al. All-optical machine learning using diffractive deep neural networks // Science. 2018. V. 361. P. 1004–1008. https://doi.org/10.1126/science.aat8084

11. Li B., Zhu Y., Fei J., et al. Multi-functional broadband diffractive neural network with a single spatial light modulator // APL Photonics. 2025. V. 10. № 1. P. 016115. https://doi.org/10.1063/5.0245832

12. Zhang Z., Feng F., Gan J., et al. Space‐time projection enabled ultrafast all‐optical diffractive neural network // Laser & Photonics Rev. 2024. V. 18. № 8. P. 2301367. https://doi.org/10.1002/lpor.202301367

13. Сошников Д.В., Досколович Л.Л., Бызов Е.В. Градиентный метод расчета каскадных ДОЭ и его применение в задаче классификации рукописных цифр // Компьютерная оптика. 2023. Т. 47. № 5. С. 691–701. https://doi.org/10.18287/2412-6179-CO-131

Soshnikov D.V., Doskolovich L.L., Byzov E.V. Gradient method for calculating cascaded DOEs and its application in the problem of classifying handwritten digits [in Russian] // Computer Opt. 2023. V. 47. № 5. P. 691–701. https://doi.org/10.18287/2412-6179-CO-131

14. Скиданов Р.В., Ханенко Ю.В., Морозов А.Е. и др. Дифракционные нейронные сети на основе 4F-схемы с жидкокристаллическими модуляторами света // Оптический журнал. 2025. Т. 92. № 12. С. 66–76. http://doi.org/10.17586/10235086-2025-92-12-66-76

Skidanov R.V., Khanenko Yu.V., Morozov A.E., et al. Diffraction neural networks based on a 4F-circuit with liquid crystal light modulators // J. Opt. Technol. 2025. V. 92. № 12. https://doi.org/10.1364/JOT.92.000000