Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

🚨 Accepted at the 2026 Photonic North Conference (PN 2026)

PIGNNRing

Physics-Informed Graph Neural Network for Inverse Design of Integrated Photonic Biosensors

In this work, we propose a physics-informed graph neural network (PI-GNN) framework for the inverse design of a microring resonator biosensor operating in the 1550 nm band. By representing the photonic structure as a graph and embedding resonance-based physical constraints directly into the learning objective, the model captures both structural connectivity and underlying electromagnetic principles.

If you use this code, please cite: Torabi, Y., Ekhteraei, A., & Khajezadeh, M. (2026). Physics-Informed Graph Neural Network for Inverse Design of Integrated Photonic Biosensors. ArXiv. https://arxiv.org/abs/2602.19082

Physics-Informed Graph Neural Network (PI-GNN). (b) Forward stage, consisting of a linear node embedding layer, two message-passing layers (T=2), sum aggregation, global mean pooling, and a multilayer perceptron (MLP) output head. (c) Inverse stage, combining data loss on predicted quantities with a physics-informed loss.

About

Physics-Informed Graph Neural Network for Inverse Design of Integrated Photonic Biosensors

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages