Training deep neural networks for the inverse design of nanophotonic structures

Abstract

Data inconsistency leads to a slow training process when deep neural networks are used for the inverse design of photonic devices, an issue that arises from the fundamental property of non-uniqueness in all inverse scattering problems. Here we show that by combining forward modeling and inverse design in a tandem architecture, one can overcome this fundamental issue, allowing deep neural networks to be effectively trained by data sets that contain non-unique electromagnetic scattering instances. This paves the way for using deep neural networks to design complex photonic structures that requires large training sets.

References in corpus (3)

• Deep Learning with Coherent Nanophotonic Circuits
• Inverse design and demonstration of a compact and broadband on-chip wavelength demultiplexer
• Trainable and Dynamic Computing: Error Backpropagation through Physical Media

Cited by in corpus (131)

• A Generative Model for Inverse Design of Metamaterials
• Deep neural networks for the evaluation and design of photonic devices
• Deep learning in nano-photonics: inverse design and beyond
• Global optimization of dielectric metasurfaces using a physics-driven neural network
• Machine learning for electronically excited states of molecules
• Freeform Diffractive Metagrating Design Based on Generative Adversarial Networks
• Intelligent Nanophotonics: Merging Photonics and Artificial Intelligence at the Nanoscale
• Design of Task-Specific Optical Systems Using Broadband Diffractive Neural Networks
• Inverse design of large-area metasurfaces
• Deep learning meets nanophotonics: A generalized accurate predictor for near fields and far fields of arbitrary 3D nanostructures
• Optimization of photonic crystal nanocavities based on deep learning
• Data-Driven Design for Metamaterials and Multiscale Systems: A Review
• Deep learning approach based on dimensionality reduction for designing electromagnetic nanostructures
• Terahertz Pulse Shaping Using Diffractive Surfaces
• Computational Imaging Without a Computer: Seeing Through Random Diffusers at the Speed of Light
• Efficient spectrum prediction and inverse design for plasmonic waveguide systems based on artificial neural networks
• Meta-optics for spatial optical analog computing
• Simulator-based training of generative models for the inverse design of metasurfaces
• Misalignment Resilient Diffractive Optical Networks
• Minireview on Disordered Optical Metasurfaces
• Pushing the limits of optical information storage using deep learning
• Enhanced Light-Matter Interactions in Dielectric Nanostructures via Machine Learning Approach
• Optical coherent dot-product chip for sophisticated deep learning regression
• Deep Learning Reveals Underlying Physics of Light-matter Interactions in Nanophotonic Devices
• Unsupervised Manifold Clustering of Topological Phononics
• Elucidating the Behavior of Nanophotonic Structures Through Explainable Machine Learning Algorithms
• Active learning of deep surrogates for PDEs: Application to metasurface design
• Plasmonic colours predicted by deep learning
• Massively Parallel Universal Linear Transformations using a Wavelength-Multiplexed Diffractive Optical Network
• Scale-, shift- and rotation-invariant diffractive optical networks
• Designing Silicon Photonic Devices using Artificial Neural Networks
• Global Inverse Design Across Multiple Photonic Structure Classes Using Generative Deep Learning
• Machine learning-assisted design of material properties
• Deep learning enabled design of complex transmission matrices for universal optical components
• Diffusion Probabilistic Model Based Accurate and High-Degree-of-Freedom Metasurface Inverse Design
• Deterministic and probabilistic deep learning models for inverse design of broadband acoustic cloak
• A Physics Based Approach for Neural Networks Enabled Design of All-Dielectric Metasurfaces
• A newcomer's guide to deep learning for inverse design in nano-photonics
• General Inverse Design of Thin-Film Metamaterials With Convolutional Neural Networks
• Iterative optimization of photonic crystal nanocavity designs by using deep neural networks
• Machine learning and excited-state molecular dynamics
• MetaNet: A new paradigm for data sharing in photonics research
• Multi-objective and categorical global optimization of photonic structures based on ResNet generative neural networks
• Mapping the Design Space of Photonic Topological States via Deep Learning
• Real-time 3D Nanoscale Coherent Imaging via Physics-aware Deep Learning
• Machine learning and evolutionary algorithm studies of graphene metamaterials for optimized plasmon-induced transparency
• Manifold Learning for Knowledge Discovery and Intelligent Inverse Design of Photonic Nanostructures: Breaking the Geometric Complexity
• Deep neural networks for the prediction of the optical properties and the free-form inverse design of metamaterials
• Inverse Design of Composite Metal Oxide Optical Materials based on Deep Transfer Learning
• Fast Design of Plasmonic Metasurfaces Enabled by Deep Learning
• Deep learning enabled strategies for modelling of complex aperiodic plasmonic metasurfaces of arbitrary size
• Probability-Density-Based Deep Learning Paradigm for the Fuzzy Design of Functional Metastructures
• Neural network design of multilayer metamaterial for temporal differentiation
• Switchable induced-transmission filters enabled by vanadium dioxide
• Efficient inverse design and spectrum prediction for nanophotonic devices based on deep recurrent neural networks
• Deep learning-based design of broadband GHz complex and random metasurfaces
• Parameterized Reinforcement Learning for Optical System Optimization
• Deep-Learning-Enabled Inverse Engineering of Multi-Wavelength Invisibility-to-Superscattering Switching with Phase-Change Materials
• Inverse design of nano-photonic wavelength demultiplexer with a deep neural network approach
• Machine learning identifies scale-free properties in disordered materials
• Heuristic Methods and Performance Bounds for Photonic Design
• Multifunctional Metasurface Design with a Generative Adversarial Network
• Deep Learning Methods for Colloidal Silver Nanoparticle Concentration and Size Distribution Determination from UV-Vis Extinction Spectra
• Over 30,000-fold field enhancement of terahertz nanoresonators enabled by rapid inverse design
• Machine prediction of topological transitions in photonic crystals
• Hybridization of Singular Plasmons via Transformation Optics
• Inverse design with flexible design targets via deep learning: Tailoring of electric and magnetic multipole scattering from nano-spheres
• Generative Inverse Design of Metamaterials with Functional Responses by Interpretable Learning
• Benchmarking deep inverse models over time, and the neural-adjoint method
• Data driven approaches in nanophotonics: A review of AI-enabled metadevices
• Constrained tandem neural network assisted inverse design of metasurfaces for microwave absorption
• Deep Neural Network Inverse Design of Integrated Nanophotonic Devices
• A General Machine Learning-based Approach for Inverse Design of One-dimensional Photonic Crystals Toward Targeted Visible Light Reflection Spectrum
• Designing thermal radiation metamaterials via hybrid adversarial autoencoder and Bayesian optimization
• A Freeform Dielectric Metasurface Modeling Approach Based on Deep Neural Networks
• Machine Learning enables Design of On-chip Integrated Silicon T-junctions with footprint of 1.2 $\mu$m x 1.2 $\mu$m
• Machine-Learning-Assisted Photonic Device Development: A Multiscale Approach from Theory to Characterization
• Active Physics Informed Deep Learning: Surrogate Modeling for Non Planar Wavefront Excitation of Topological Nanophotonic Devices
• Integration of Programmable Diffraction with Digital Neural Networks
• Deep learning for nano-photonic materials -- The solution to everything!?
• Predictions of Transient Vector Solution Fields with Sequential Deep Operator Network
• A Novel Modeling Approach for All-Dielectric Metasurfaces Using Deep Neural Networks
• Acoustic Structure Inverse Design and Optimization Using Deep Learning
• Chat to Chip: Large Language Model Based Design of Arbitrarily Shaped Metasurfaces
• Design of ultracompact broadband focusing spectrometers based on deep diffractive neural networks
• Deep Convolutional Neural Networks to Predict Mutual Coupling Effects in Metasurfaces
• Transfer learning-assisted inverse modeling in nanophotonics based on mixture density networks
• MaxwellNet: Physics-driven deep neural network training based on Maxwell's equations
• Finding the best design parameters for optical nanostructures using reinforcement learning
• Machine Learning-Based Optimization of Chiral Photonic Nanostructures: Evolution- and Neural Network-Based Design
• Compounding meta-atoms into meta-molecules with hybrid artificial intelligence techniques
• Predicting VCSEL Emission Properties Using Transformer Neural Networks
• MCP-Enabled LLM for Meta-optics Inverse Design: Leveraging Differentiable Solver without LLM Expertise
• Migrating Knowledge between Physical Scenarios based on Artificial Neural Networks
• Fast multi-channel inverse design through augmented partial factorization
• Fixed-Attention Mechanism for Deep-Learning-Assisted Design of High-Degree-of-Freedom 3D Metamaterials
• Spectral domain inverse design for accelerating nanocomposite metamaterials discovery
• Multifunctional Meta-Optic Systems: Inversely Designed with Artificial Intelligence
• Analysis and design of transition radiation in layered uniaxial crystals using Tandem neural networks
• Automated Optical Multi-layer Design via Deep Reinforcement Learning
• A novel guided deep learning algorithm to design low-cost SPP films
• Multiplexed Supercell Metasurface Design and Optimization with Tandem Residual Networks
• Neural-adjoint method for the inverse design of all-dielectric metasurfaces
• Deep learning for the modeling and inverse design of radiative heat transfer
• Deep Learning to Automate Parameter Extraction and Model Fitting of Two-Dimensional Transistors
• Probabilistic representation and inverse design of metamaterials based on a deep generative model with semi-supervised learning strategy
• Mixed Integer Neural Inverse Design
• Mixture Manifold Networks: A Computationally Efficient Baseline for Inverse Modeling
• Mode visualisation and control of complex lasers using neural networks
• TAEN: A Model-Constrained Tikhonov Autoencoder Network for Forward and Inverse Problems
• Physical limits on electromagnetic response
• Inverse design of multilayer nanoparticles using artificial neural networks and genetic algorithm
• Generative Deep Learning Model for a Multi-level Nano-Optic Broadband Power Splitter
• Deep learning estimation of complex reverberant wave fields by a programmable metasurface
• PyMieDiff: A differentiable Mie scattering library
• Programmable Multifunctional Plasmonic Waveguide System based on Coding Metamaterials and Inverse Design
• Poles of the Scattering Matrix: An Inverse Method for Designing Photonic Resonators
• Magnetic and electric Purcell factor control through geometry optimization of high index dielectric nanostructures
• High-Freedom Inverse Design with Deep Neural Network for Metasurface Filter in the Visible
• Broadband vectorial ultra-flat optics with experimental efficiency up to 99% in the visible via universal approximators
• Numerical study of laser micro- and nano-processing of nanocomposite porous materials
• A Reinforcement learning method for Optical Thin-Film Design
• Intelligent Metasurface Imager and Recognizer
• A general framework for knowledge integration in machine learning for electromagnetic scattering using quasinormal modes
• A Quantum-Inspired Probabilistic Model for the Inverse Design of Meta-Structures
• Spectra2pix: Generating Nanostructure Images from Spectra
• Wavelength Controllable Forward Prediction and Inverse Design of Nanophotonic Devices Using Deep Learning
• Towards a universal meta-optics solver via large language models
• Ultra-fast and accurate multimode waveguide design based on dataset-based eigenmode expansion method
• Spectral splitting and concentration of broadband light using neural networks
• Ultra low loss broadband 1 $\times$ 2 optical power splitters with various splitting ratios