Geometry-Aware Neural Optimizer for Shape Optimization and Inversion

TL;DR

GANO is a geometry-aware neural optimizer that unifies shape representation, prediction, and optimization in a differentiable framework, enabling efficient shape inversion and optimization with state-of-the-art results.

Contribution

It introduces GANO, an end-to-end differentiable system that stabilizes shape updates, supports part-wise control, and accelerates geometry processing for PDE-based shape optimization.

Findings

Achieves up to +55.9% lift-to-drag ratio improvement for airfoils.

Reduces drag by approximately 7% for 3D vehicles.

Demonstrates state-of-the-art accuracy and stable, controllable shape updates.

Abstract

Geometry is central to PDE-governed systems, motivating shape optimization and inversion. Classical pipelines conduct costly forward simulation with geometry processing, requiring substantial expert effort. Neural surrogates accelerate forward analysis but do not close the loop because gradients from objectives to geometry are often unavailable. Existing differentiable methods either rely on restrictive parameterizations or unstable latent optimization driven by scalar objectives, limiting interpretability and part-wise control. To address these challenges, we propose Geometry-Aware Neural Optimizer (\textbf{\textsc{GANO}}), an end-to-end differentiable framework that unifies geometry representation, field-level prediction, and automated optimization/inversion in a single latent-space loop. \textsc{GANO} encodes shapes with an auto-decoder and stabilizes latent updates via a denoising mechanism, and a geometry-informed surrogate provides a reliable gradient pathway for geometry updates. Moreover, \textsc{GANO} supports part-wise control through null-space projection and uses remeshing-free projection to accelerate geometry processing. We further prove that denoising induces an implicit Jacobian regularization that reduces decoder sensitivity, yielding controlled deformations. Experiments on three benchmarks spanning 2D Helmholtz, 2D airfoil, and 3D vehicles show state-of-the-art accuracy and stable, controllable updates, achieving up to +55.9% lift-to-drag improvement for airfoils and ~7% drag reduction for vehicles.