Abstract:The research of metamaterials has achieved enormous success in the manipulation of light in an artificially prescribed manner using delicately designed sub-wavelength structures, so-called meta-atoms. Even though modern numerical methods allow to accurately calculate the optical response of complex structures, the inverse design of metamaterials is still a challenging task due to the non-intuitive and non-unique relationship between physical structures and optical responses. To better unveil this implicit relationship and thus facilitate metamaterial design, we propose to represent metamaterials and model the inverse design problem in a probabilistically generative manner. By employing an encoder-decoder configuration, our deep generative model compresses the meta-atom design and optical response into a latent space, where similar designs and similar optical responses are automatically clustered together. Therefore, by sampling in the latent space, the stochastic latent variables function as codes, from which the candidate designs are generated upon given requirements in a decoding process. With the effective latent representation of metamaterials, we can elegantly model the complex structure-performance relationship in an interpretable way, and solve the one-to-many mapping issue that is intractable in a deterministic model. Moreover, to alleviate the burden of numerical calculation in data collection, we develop a semi-supervised learning strategy that allows our model to utilize unlabeled data in addition to labeled data during training, simultaneously optimizing the generative inverse design and deterministic forward prediction in an end-to-end manner. On a data-driven basis, the proposed model can serve as a comprehensive and efficient tool that accelerates the design, characterization and even new discovery in the research domain of metamaterials and photonics in general.

MagGen: A Graph-Aided Deep Generative Model for Inverse Design of Permanent Magnets

MagGen: A graph aided deep generative model for inverse design of stable, permanent magnets

VAE for Modified 1-Hot Generative Materials Modeling, A Step Towards Inverse Material Design

Deep generative modeling for mechanistic-based learning and design of metamaterial systems

Predicting magnetic properties of van der Waals magnets using graph neural networks

Image-based prediction and optimization of hysteresis properties of nanocrystalline permanent magnets using deep learning

Generative deep learning for the inverse design of materials

Revolutionising inverse design of magnesium alloys through generative adversarial networks

Generative Deep Neural Networks for Inverse Materials Design Using Backpropagation and Active Learning

Data-efficient and Interpretable Inverse Materials Design using a Disentangled Variational Autoencoder

Physics-informed machine learning combining experiment and simulation for the design of neodymium-iron-boron permanent magnets with reduced critical-elements content

Probabilistic representation and inverse design of metamaterials based on a deep generative model with semi-supervised learning strategy

Topological magnetic structure generation using VAE-GAN hybrid model and discriminator-driven latent sampling

A Generative Model for Accelerated Inverse Modelling Using a Novel Embedding for Continuous Variables

Deep Generative Models for Materials Discovery and Machine Learning-Accelerated Innovation

Magnetic Anomaly Inversion Based on Convolutional Neural Network

Inverse design machine learning model for metallic glasses with good glass-forming ability and properties

Advancing magnetic material discovery through machine learning: Unveiling new manganese-based materials

Generative Adversarial Networks for Inverse Design of Two-Dimensional Spinodoid Metamaterials

A Generative Model for Inverse Design of Metamaterials

Generative deep learning as a tool for inverse design of high-entropy refractory alloys