paper-with-me

홈 › Papers

Challenges in Training PINNs: A Loss Landscape Perspective

2024-02-02 · Pratik Rathore, Weimu Lei, Zachary Frangella, Lu Lu, Madeleine Udell

This paper explores challenges in training Physics-Informed Neural Networks (PINNs), emphasizing the role of the loss landscape in the training process. We examine difficulties in minimizing the PINN loss function, particularly due to ill-conditioning caused by differential operators in the residual term. We compare gradient-based optimizers Adam, L-BFGS, and their combination Adam+L-BFGS, showing the superiority of Adam+L-BFGS, and introduce a novel second-order optimizer, NysNewton-CG (NNCG), which significantly improves PINN performance. Theoretically, our work elucidates the connection between ill-conditioned differential operators and ill-conditioning in the PINN loss and shows the benefits of combining first- and second-order optimization methods. Our work presents valuable insights and more powerful optimization strategies for training PINNs, which could improve the utility of PINNs for solving difficult partial differential equations.

📄 PDF Abstract BibTeX arXiv:2402.01868

Code (1)

pratikrathore8/opt_for_pinns 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Adam 설명 없음

Similar Papers 제목 키워드 기반

Loss Landscape Engineering via Data Regulation on PINNs

2022-05-16 · Vignesh Gopakumar, Stanislas Pamela, Debasmita Samaddar

Physics-Informed Neural Networks have shown unique utility in parameterising the solution of a well-defined partial differential equation using automatic differentiation and residual losses. Though they provide theoretic…

A unified scalable framework for causal sweeping strategies for Physics-Informed Neural Networks (PINNs) and their temporal decompositions

2023-02-28 · Michael Penwarden, Ameya D. Jagtap, Shandian Zhe, George Em Karniadakis 외

Physics-informed neural networks (PINNs) as a means of solving partial differential equations (PDE) have garnered much attention in the Computational Science and Engineering (CS&E) world. However, a recent topic of inter…

Transfer Learning

iPINNs: Incremental learning for Physics-informed neural networks

2023-04-10 · Aleksandr Dekhovich, Marcel H. F. Sluiter, David M. J. Tax, Miguel A. Bessa

Physics-informed neural networks (PINNs) have recently become a powerful tool for solving partial differential equations (PDEs). However, finding a set of neural network parameters that lead to fulfilling a PDE can be ch…

Incremental LearningMulti-Task LearningTransfer Learning

Taming the Loss Landscape of PINNs with Noisy Feynman-Kac Supervision: Operator Preconditioning and Non-Asymptotic Error Bounds

2026-05-30 · Nathanael Tepakbong, Hanyu Hu, Chengyu Liu, Xiang Zhou arxiv

Physics-Informed Neural Networks (PINNs) often train slowly or fail to converge on challenging partial differential equations (PDEs), a behavior recently linked to severely ill-conditioned loss landscapes inherited from …

Mitigating Gradient Pathology in PINNs through Aligned Constraint

2026-05-24 · Yichen Luo, Peiyu Zhu, Dongxiao Hu, Jia Wang 외 arxiv

While Physics-Informed Neural Networks (PINNs) are powerful for solving Partial Differential Equations (PDEs), their training is often paralyzed by gradient pathology. The gradients from the PDE residuals and boundary co…