Optical multi-task learning using a multifunctional diffractive processor with “learned” structured illumination
Optics Letters · 15.04.2026 · 51 (8) · 2236
DOI: 10.1364/ol.591818 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Letters, online erschienen am 13.04.2026. Verfasst von Jiajun Zhang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 12 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Special Fund for Key Basic Research in Jiangsu Province, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
Diffractive deep neural network (D 2 NN), as a novel optical machine learning framework, has attracted much attention due to its superior advantages in inference speed and power consumption. Here, we demonstrate a new, to the best of our knowledge, framework design for multifunctional D 2 NN with “learned” structured illumination, incorporating the illumination patterns as trainable parameters into the network optimization process. Compared with the traditional D 2 NN design, this network framework enables the reuse of diffractive layers while quickly switching between different machine learning tasks, significantly expanding the network’s practicality. Based on this, two different training approaches are proposed: one based on transfer learning and the other based on constructing a multi-input and multi-output (MIMO) neural network. The approaches are tested on three different tasks: MNIST dataset classification, Fashion-MNIST dataset classification, and computational imaging. The D 2 NN framework we proposed not only improves the flexibility of optical neural networks but also paves a new path for the construction of multifunctional optical machine learning platforms.
Förderung
- National Natural Science Foundation of China
- Special Fund for Key Basic Research in Jiangsu Province
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Zhang, J., Wen, K., Wang, Y., Wang, X., Li, J., & Lu, W. (2026). Optical multi-task learning using a multifunctional diffractive processor with “learned” structured illumination. Optics Letters, 51(8), 2236. https://doi.org/10.1364/ol.591818
Vancouver
Zhang J, Wen K, Wang Y, Wang X, Li J, Lu W. Optical multi-task learning using a multifunctional diffractive processor with “learned” structured illumination. Optics Letters. 2026;51(8):2236. doi:10.1364/ol.591818
BibTeX
@article{Zhang2026,
author = {Zhang, Jiajun and Wen, Kexin and Wang, Yishun and Wang, Xiang and Li, Jiaqi and Lu, Weibing},
title = {Optical multi-task learning using a multifunctional diffractive processor with “learned” structured illumination},
journal = {Optics Letters},
year = {2026},
volume = {51},
number = {8},
pages = {2236},
doi = {10.1364/ol.591818}
}
EUVIMED