Introduces ∂BPM volumetric training for visible-range diffractive neural networks, reducing design-to-device mismatch and raising FDTD-validated classification accuracy from 50% to 90%.
All-optical machine learning using diffractive deep neural networks
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
A surrogate matrix-space training stage followed by adjoint-based physical realization allows scalable design of photonic neural networks that match ideal accuracy on image tasks with orders-of-magnitude fewer electromagnetic simulations.
A double pass through a digital micromirror device adds a quadratic nonlinearity to an optical reservoir computer using only linear optics, improving Mackey-Glass chaotic forecasting on one experimental apparatus.
The roadmap reviews advances in generating and manipulating high-dimensional quantum states of light across photonic degrees of freedom and outlines common challenges for their integration into future quantum technologies.
citing papers explorer
-
Beyond the Thin-Layer Limit: Differentiable Volumetric Training for Visible-Range Diffractive Neural Networks
Introduces ∂BPM volumetric training for visible-range diffractive neural networks, reducing design-to-device mismatch and raising FDTD-validated classification accuracy from 50% to 90%.
-
Scalable Photonic Neural Networks via Surrogate Scattering-Matrix Inverse Design
A surrogate matrix-space training stage followed by adjoint-based physical realization allows scalable design of photonic neural networks that match ideal accuracy on image tasks with orders-of-magnitude fewer electromagnetic simulations.
-
Optical Reservoir Computing with Structural Nonlinearity for Forecasting Chaotic Time Series
A double pass through a digital micromirror device adds a quadratic nonlinearity to an optical reservoir computer using only linear optics, improving Mackey-Glass chaotic forecasting on one experimental apparatus.
-
High-Dimensional Quantum Photonics: Roadmap
The roadmap reviews advances in generating and manipulating high-dimensional quantum states of light across photonic degrees of freedom and outlines common challenges for their integration into future quantum technologies.