Learning Data-Efficient and Generalizable Neural Operators via Fundamental Physics Knowledge

ICLR 2026 Conference Submission14708 Authors

Keywords: Neural Operator, PDE, Fundamental Physics Knowledge

TL;DR: We propose to incorporate fundamental physics knowledge into learning neural operators to enhance its data efficiency, long-term consistency, and OOD generalization.

Abstract:

Recent advances in scientific machine learning (SciML) have enabled neural operators (NOs) to serve as powerful surrogates for modeling the dynamic evolution of physical systems governed by partial differential equations (PDEs). While existing approaches focus primarily on learning simulations from the target PDE, they often overlook more fundamental physical principles underlying these equations. Inspired by how numerical solvers are compatible with simulations of different settings of PDEs, we propose a multiphysics training framework that jointly learns from both the original PDEs and their simplified basic forms. Our framework enhances data efficiency, reduces predictive errors, and improves out-of-distribution (OOD) generalization, particularly in scenarios involving shifts of physical parameters and synthetic-to-real transfer. Our method is architecture-agnostic and demonstrates consistent improvements in normalized root mean square error (nRMSE) across a wide range of 1D/2D/3D PDE problems. Through extensive experiments, we show that explicit incorporation of fundamental physics knowledge significantly strengthens the generalization ability of neural operators.
We promise to release models and data upon acceptance.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Submission Number: 14708