Full-Vector Diffractive Deep Neural Networks
Abstract
Conventional diffractive deep neural networks (D2NNs) treat polarization as independent channels and rely on scalar or semi-vectorial propagation models, which neglect cross polarization coupling and vector diffraction. Here we propose a vector D2NN (V-D2NN) that embeds the full vector angular spectrum method into the end-to-end training pipeline. This physically rigorous framework describes vectorial light-matter interactions across cascaded diffractive layers, enabling direct optimization of polarization conversion, spin-orbit coupling, and vectorial interference without external polarization optics. We demonstrate the V-D2NN on polarization-multiplexed tasks--ector beam generation, polarization dependent imaging and classification, and multiple channel optical encryption--where it consistently outperforms scalar and semi-vectorial counterparts. In longitudinal-field engineering, the V-D2NN actively shapes a prescribed longitudinal field with a normalized correlation of 0.839, a capability inaccessible to scalar-propagation models. An open-source training framework is also provided to support further development of vectorial diffractive optics for computing, sensing, and communications.
Related
- Deep diffractive optical neural networks for detecting skyrmionic topologies of light and their live training
- DNNsolver: Accurate and Efficient Deep Learning Modeling of Polarization-Sensitive Diffractive Neural Networks
- Beyond the Thin-Layer Limit: Differentiable Volumetric Training for Visible-Range Diffractive Neural Networks
- Edge-enhanced diffractive neural networks based on spin-multiplexed nonlocal metasurfaces
- A graph neural network for deep learning-based full event interpretation at FCC-ee
BibTeX
@misc{zhao2026fullvector,
title = {{Full-Vector Diffractive Deep Neural Networks}},
author = {Shengmei Zhao and Gongyuan Wang and Zhiwan Wang and Jianan Shao and Haolin Liu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.11274},
eprint = {2610.11274},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.11274},
}
RIS
TY - UNPB
TI - Full-Vector Diffractive Deep Neural Networks
AU - Shengmei Zhao
AU - Gongyuan Wang
AU - Zhiwan Wang
AU - Jianan Shao
AU - Haolin Liu
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.11274
UR - https://arxiv.org/abs/2610.11274
AB - Conventional diffractive deep neural networks (D2NNs) treat polarization as independent channels and rely on scalar or semi-vectorial propagation models, which neglect cross polarization coupling and vector diffraction. Here we propose a vector D2NN (V-D2NN) that embeds the full vector angular spectrum method into the end-to-end training pipeline. This physically rigorous framework describes vectorial light-matter interactions across cascaded diffractive layers, enabling direct optimization of polarization conversion, spin-orbit coupling, and vectorial interference without external polarization optics. We demonstrate the V-D2NN on polarization-multiplexed tasks--ector beam generation, polarization dependent imaging and classification, and multiple channel optical encryption--where it consistently outperforms scalar and semi-vectorial counterparts. In longitudinal-field engineering, the V-D2NN actively shapes a prescribed longitudinal field with a normalized correlation of 0.839, a capability inaccessible to scalar-propagation models. An open-source training framework is also provided to support further development of vectorial diffractive optics for computing, sensing, and communications.
ER -