Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 11, 2026

Fabrication of 1-D Photonic Crystal Cavity on a Nanofiber Using Femtosecond Laser-induced Ablation

Fabrication of 1-D Photonic Crystal Cavity on a Nanofiber Using Femtosecond Laser-induced Ablation

Published on: February 25, 2017

Efficient nanophotonic devices optimization using deep neural network trained with physics-based transfer learning

Gibaek Kim¹, Jungho Kim²

¹Department of Information Display, Kyung Hee University, 26 Kyungheedae-ro, Dongdaemun-gu, Seoul, 02447, South Korea.

Scientific Reports

|November 13, 2025

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Cd99l2 regulates excitatory synapse development and restrains immediate-early gene activation.

Cell reports·2026

Same author

Electrochemical-sensor-assisted lab-in-a-cartridge (EC-LIC) for on-site detection of SARS-CoV-2 with a self-contained heating system.

Biosensors & bioelectronics·2026

Same author

Integrated analysis of miRNA and transcription factor gene expression profiles associated with anti-tuberculosis treatment responses.

BMC infectious diseases·2026

Same author

Proteolytic cleavage of G3BP1 by calpain 1 couples NMDAR activation to mTOR-dependent local translation.

EMBO reports·2026

Same author

Intertwined orders in a quantum-entangled metal.

Nature materials·2026

Same author

AVN944 Elicits Apoptotic Responses and Impedes Tumorigenic Potential in Ewing's Sarcoma Cells.

International journal of biological sciences·2026

Same journal

Therapeutic potential of crude protein extracts from two Egyptian freshwater snails Lanistes carinatus and Bellamya unicolor.

Scientific reports·2026

Same journal

Microbial contamination of donor corneas and post-keratoplasty endophthalmitis: a comparison between Japanese and U.S. eye banks using cold storage.

Scientific reports·2026

Same journal

Prevalence and contributing factors of virological non-suppression among adult patients on first-line antiretroviral therapy in tertiary hospitals in Ethiopia.

Scientific reports·2026

Same journal

An in vitro comparison of color stability between alkasite and different restorative materials in various staining solutions.

Scientific reports·2026

Same journal

Toward accessible mRNA LNP formulation: systematic evaluation of mixing strategies and key parameters.

Scientific reports·2026

Same journal

A network analysis of personality traits, mentalizing, and psychological health in Chinese college students.

Scientific reports·2026

See all related articles

We developed a neural network framework for photonic device optimization, significantly reducing data needs and computation time. This approach accelerates the design of devices like quantum cascade lasers by over 80,000 times.

Area of Science:

Photonics
Computational Physics
Machine Learning

Background:

Photonic device optimization is crucial but hindered by high data generation costs and imbalanced feature importance.
Existing methods struggle with complex design spaces and multiple local optima, demanding significant computational resources.
Efficient design automation for photonic devices, such as quantum cascade lasers (QCLs), requires novel approaches to overcome these limitations.

Purpose of the Study:

To propose a generalizable neural network (NN)-based surrogate modeling framework for efficient photonic device optimization.
To address challenges of imbalanced feature importance, high data costs, and multiple local optima in photonic design.
To accelerate the design automation process for complex photonic devices using minimal data resources.

Keywords:

Physics based neural network Transfer learning mid-IR QCLs

Related Experiment Videos

Last Updated: Jan 11, 2026

Fabrication of 1-D Photonic Crystal Cavity on a Nanofiber Using Femtosecond Laser-induced Ablation

Fabrication of 1-D Photonic Crystal Cavity on a Nanofiber Using Femtosecond Laser-induced Ablation

Published on: February 25, 2017

Main Methods:

Developed a framework combining physics-based transfer learning (PBTL)-enhanced surrogate modeling with multi-objective genetic algorithms (GAs).
Integrated a deep neural network total predictor (DNN-TP) with GAs to replace computationally expensive numerical simulations.
Utilized PBTL to transfer knowledge from a core predictor (DNN-CP) trained on specific device regions to improve generalization.

Main Results:

Achieved an optimization speed-up of over 80,000 times by replacing numerical simulations with the DNN-TP model.
PBTL improved prediction accuracy by 0.69%, reducing training data requirements by 50%.
Generated more feasible device structures, showing a 60% improvement in evaluation metrics during optimization.

Conclusions:

The proposed NN-based surrogate modeling framework offers a highly efficient and generalizable solution for photonic device optimization.
The integration of PBTL and DNN-TP with GAs effectively addresses data scarcity and computational cost challenges.
This framework significantly accelerates the exploration of the design space, enabling large-scale optimization for devices like QCLs.