Inverse design method of optical devices for generating target near-fields via physics-informed neural networks
Optics Letters
|February 27, 2026
Summary
We introduce a new method using physics-informed neural networks (PINNs) for optical device inverse design. This approach generates target optical near-fields efficiently without needing precomputed data or adjoint gradients.
Area of Science:
- Computational electromagnetics
- Optical device design
- Physics-informed machine learning
Background:
- Traditional inverse design methods for optical devices often require extensive training data or computationally expensive adjoint methods.
- Generating specific optical near-field distributions is crucial for applications like nanophotonics and advanced imaging.
- Existing approaches may struggle with complex boundary conditions and realizable material constraints.
Purpose of the Study:
- To develop a novel inverse design framework for optical devices capable of generating user-defined near-field patterns.
- To leverage physics-informed neural networks (PINNs) to solve the inverse design problem efficiently and accurately.
- To enable the discovery of non-intuitive optical structures for targeted electromagnetic field manipulation.
Main Methods:
- Embedding the finite-difference Helmholtz equation directly into the PINN loss function to ensure wave equation compliance.
- Utilizing a data-loss function to enforce the generation of prescribed near-field distributions.
- Incorporating penalty and binarization losses to constrain the device's refractive index to physically realizable values.
Main Results:
- The PINN-based method accurately predicts optical fields, validated against generalized Lorenz-Mie theory and full-wave simulations.
- The framework successfully designed optical devices for generating target near-fields, including photonic nanojets and photonic hooks.
- The method demonstrated the ability to discover complex, non-intuitive device structures automatically.
Conclusions:
- Physics-informed neural networks offer a powerful and data-efficient approach for the inverse design of optical devices.
- The proposed method eliminates the need for precomputed training datasets and problem-specific adjoint gradients.
- This framework provides a versatile tool for designing optical elements with arbitrary near-field characteristics.
Related Concept Videos
01:24Vision
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
01:16Focusing of Light in the Eye
Light rays enter the eye through the cornea, a transparent dome-shaped tissue that is the eye's outermost layer. The cornea bends or refracts, light rays traveling to the pupil. The shape of the cornea determines how much of the light is bent and whether the image will be focused correctly on the retina at the back of the eye. Once the light has passed through both refraction layers, it converges into a single focal point onto a small area. This is where photoreceptors start transforming...
01:30Electronic Distance Measuring Instruments
Electronic Distance Measuring Instruments (EDMs) are essential tools in modern surveying, offering precise distance measurements by emitting electromagnetic signals and calculating the time required for these signals to travel to a target and return. Two primary types of signals are used in EDMs — light waves and microwaves — each suited to specific environmental and distance requirements. Light-wave-based EDMs utilize either infrared or laser light, providing high accuracy over short distances...