AutoTandemML: Active Learning Enhanced Tandem Neural Networks for Inverse Design Problems

TL;DR

This paper introduces AutoTandemML, a hybrid method combining active learning with Tandem Neural Networks to improve inverse design efficiency and accuracy in high-dimensional, complex design spaces.

Contribution

It presents a novel hybrid approach that leverages active learning with Tandem Neural Networks for more efficient inverse design, reducing data requirements while maintaining accuracy.

Findings

Outperforms standard methods on benchmark problems

Achieves higher accuracy with fewer training samples

Effective in high-dimensional inverse design tasks

Abstract

Inverse design in science and engineering involves determining optimal design parameters that achieve desired performance outcomes, a process often hindered by the complexity and high dimensionality of design spaces, leading to significant computational costs. To tackle this challenge, we propose a novel hybrid approach that combines active learning with Tandem Neural Networks to enhance the efficiency and effectiveness of solving inverse design problems. Active learning allows to selectively sample the most informative data points, reducing the required dataset size without compromising accuracy. We investigate this approach using three benchmark problems: airfoil inverse design, photonic surface inverse design, and scalar boundary condition reconstruction in diffusion partial differential equations. We demonstrate that integrating active learning with Tandem Neural Networks outperforms standard approaches across the benchmark suite, achieving better accuracy with fewer training samples.