NeCLO: Neural Convolutional Learning Optimizer for Electromagnetics

Zhang Yanxin, Zaifeng Yang, Xinyu Yang, Yueming Lyu

Submission Type: I want my submission to be considered for both oral and poster presentation.

Keywords: FDTD Solver, Computational Electromagnetics, Physics Foundation Model, Scientific Computing, Physics-Informed Deep Learning

TL;DR: We introduce NeCLO (Neural Convolutional Learning Optimizer), a hardware-accelerated differentiable solver that maps FDTD to 3D convolutions, achieving bit-exact numerical fidelity for accurate, physics-driven inverse design.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 289