Neural Operator-based Curriculum Learning for Physics-Informed Neural Networks

ICLR 2026 Conference Submission15858 Authors

Keywords: Curriculum Learning, PINNs, Nerual Operator, PDEs

Abstract: In this paper, we tackle the critical failure modes of Physics-Informed Neural Networks (PINNs), such as spectral bias and ill-conditioning, which lead to poor convergence on complex PDEs. We identify two key shortcomings in existing curriculum learning methods for PINNs: unreliable knowledge transfer between stages and a reliance on manual, ad-hoc curriculum design. To overcome these limitations, we present Neural Operator-based Curriculum Learning (NOCL), a unified framework that leverages Neural Tangent Kernel (NTK) theory to automate curriculum generation and employs neural operators to enable robust, dynamic knowledge transfer across curriculum stages. By dynamically training the operator and filtering data for PINN initialization, our approach ensures scalable and effective learning across progressively difficult tasks. Experiments verify that our proposed NOCL achieves state-of-the-art performance, markedly improving convergence and generalization over existing methods.

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

Submission Number: 15858