Optical metasurfaces enable devices to interact with light in unique ways by modulating phase, polarization, or intensity. A metasurface, composed of individual subwavelength scatterers known as meta-atoms, can be designed to provide unparalleled control of wavefronts for a variety of optical applications, yet the design of such devices is often unintuitive and challenging due to computationally expensive forward simulations and the number of free parameters. To overcome this, there is interest in developing inverse design methods as an alternative to conventional forward design. Inverse design leverages machine learning algorithms to effectively search a problem space, starting from application and resulting in solution parameters. In this work, we adopt an inverse design approach that involves targeted forward simulations of arbitrary meta-atoms. To ensure that the dataset captures all possible shapes and rotations of near field responses with second order accuracy, it is constructed using meta-atoms with varying geometries and corresponding phase shifts, including the effect of nearest neighbors. A custom deep learning system is developed to extract meaningful features from this near field response. The proposed framework provides flexibility to produce an inverse design paradigm for generalized metasurface applications without the need for repeated forward simulations. Additionally, the machine learning model is highly effective in reconstructing electric fields, irrespective of the loss function used.
The adoption of neural networks for optical component design has increased rapidly in recent years. In this design framework, the numerical simulation of optical wave propagation and material wave modulation are encoded directly as layers of a neural network. This direct encoding enables the optimization of physical quantities (e.g., the transmissivity values of the diffractive optical elements) by gradient descent and the backpropagation algorithm. For the body of work which uses these networks for simulation and optimization, there is a tendency to treat the training process as identical to traditional deep neural networks. However, to the best of our knowledge, there is yet an explicit evaluation of training parameters to support this intuition. This work aims to help fill this gap by providing an exploration and evaluation of data variety to help accelerate those in the community who wish to use this emerging design framework.The application of neural networks in optical component design has witnessed rapid growth in recent years. This design framework encodes the numerical simulation of optical wave propagation and material wave modulation directly within neural network layers, enabling the optimization of physical quantities, such as transmissivity values of diffractive optical elements, through gradient descent and backpropagation algorithms. Physics-informed neural networks have been employed in designing diffractive deep neural networks, optimizing holograms for near-eye displays and creating multi-objective traditional optics. However, there remains a lack of evaluation for training parameters, and discrete sampling considerations are often overlooked. To address these gaps, this study examines the impact of dataset variety on physics-informed neural networks in optimizing lenses that either satisfy or violate the Nyquist sampling criteria. Results show that increased data variety enhances optimized lens performance across all cases. Optimized lenses demonstrate improved imaging performance by reducing diffraction orders present in aliased analytical lenses. Moreover, we reveal that low data variety leads to overfit lenses that function as selective imagers, providing valuable insights for future lens design and optimization.
Computer-generated holography (CGH) has enabled the formation of arbitrary images through complex spatial light modulation. The optimization of spatial light modulators (SLMs) and diffractive optical elements (DOEs) is aimed to solve the well-known phase retrieval problem. This paper proposes a physically constrained artificial neural network (ANN) designed to solve the phase retrieval problem for CGH. We show that through careful selection of model structural parameters and by limiting the scope of model optimization, we can encode Fresnel Diffraction equations directly into an ANN. We train the proposed model to overfit to a single image, i.e., the model finds the SLM phase delays required to produce the desired image. The proposed model performs well with outputs that qualitatively compare well with ideal images. The method proposed in this work holds value for those who require confidence that their machine learning techniques are physically realizable.
Neural network based classifiers have been shown to suffer from image perturbations in the form of 2-dimensional transformations. These transformations lack physical constraints making them less of a practical concern and more of a theoretical interest. This paper pushes to produce 3-dimensional materials to mimic these 2-dimensional image transformations by using artificial neural networks to regress material parameters. The neural networks are trained on simulation data from full-wave simulations and physics-based ray tracing simulations. Two neural network models are developed to regress material parameters of a common transformation optics solution, and a Gaussian blur, respectively. The model trained for the transformation optics solution was able to find a unique material solution whose simulated waveform generally matches an analytical solution. The model trained for the Gaussian blur was unable to find an adequate material solution for the image transformation possibly due to the constraints placed on the regression by the ray tracing simulation. Finally, a framework is proposed to combine the ray tracing and full-wave simulations to produce more accurate data, enabling a better regression of material parameters for image transformations.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.