Building Trust in PINNs: Error Estimation through Finite Difference Methods
Department of Artificial Intelligence, Fraunhofer Heinrich Hertz Institute(人工智能部门,弗劳恩霍夫海因里希·赫兹研究所);Department of Electrical Engineering and Computer Science, Technische Universität Berlin(电气工程与计算机科学系,柏林技术大学);BIFOLD - Berlin Institute for the Foundations of Learning(BIFOLD—柏林学习与数据基础研究所);Centre of eXplainable Artificial Intelligence, Technological University Dublin(可解释人工智能中心,都柏林技术大学)
Aleksander Krasowski, René P. Klausen, Aycan Celik, Sebastian Lapuschkin, Wojciech Samek, Jonas Naujoks
Physics-informed neural networks (PINNs) constitute a flexible deep learning approach for solving partial differential equations (PDEs), which model phenomena ranging from heat conduction to quantum mechanical systems. Despite their flexibility, PINNs offer limited insight into how their predictions deviate from the true solution, hindering trust in their prediction quality. We propose a lightweight post-hoc method that addresses this gap by producing pointwise error estimates for PINN predictions, which offer a natural form of explanation for such models, identifying not just whether a prediction is wrong, but where and by how much. For linear partial differential equations, the error between a PINN approximation and the true solution satisfies the same differential operator as the original problem, but driven by the PINN's PDE residual as its source term. We solve this error equation numerically using finite difference methods requiring no knowledge of the true solution. Evaluated on several benchmark PDEs, our method yields accurate error maps at low computational cost, enabling targeted and interpretable validation of PINNs.
Physics-informed neural networks need a physicist to be accurate: the case of mass and heat transport in Fischer-Tropsch catalyst particles
SoftServe Inc.(SoftServe公司);Taras Shevchenko National University of Kyiv(基辅塔拉斯·舍甫琴科国立大学);Shell India Markets Pvt. Ltd.(壳牌印度市场私人有限公司);Shell Global Solutions International B.V.(壳牌全球解决方案国际有限公司)
Uncertainty quantification for noisy inputs-outputs in physics-informed neural networks and neural operators
Brown University(布朗大学);Huazhong University of Science and Technology(华中科技大学);Institute of Interdisciplinary Research for Mathematics and Applied Science, School of Mathematics and Statistics, Huazhong University of Science and Technology(华中科技大学数学与统计学院数学与应用科学交叉研究院)
When PINNs Go Wrong: Pseudo-Time Stepping Against Spurious Solutions
Institute for Foundations of Data Science, Yale University(数据科学基础研究所,耶鲁大学);Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania(机械工程与应用力学系,宾夕法尼亚大学);Industrial Engineering & Management Science, Northwestern University(工业工程与管理科学,西北大学)
A Numerical Method for Coupling Parameterized Physics-Informed Neural Networks and FDM for Advanced Thermal-Hydraulic System Simulation
Division of Advanced Nuclear Engineering, Pohang University of Science and Technology(浦项科技大学先进核工程学部);Mueunjae School of Undergraduate Studies, Pohang University of Science and Technology(浦项科技大学文在本科书院);Department of Quantum System Engineering, Jeonbuk National University(全北国立大学量子系统工程系)
Benchmarking Physics-Informed Neural Networks and Boundary Elements Methods for Wave Scattering
Mathematical Applications in Science and Engineering Research Group, School of Applied Sciences and Engineering, Universidad EAFIT, Medellín, Colombia(数学应用科学与工程研究组,应用科学与工程学院,埃法特大学,麦德林,哥伦比亚);Nature and Cities Research Group, School of Applied Sciences and Engineering, Universidad EAFIT, Medellín, Colombia(自然与城市研究组,应用科学与工程学院,埃法特大学,麦德林,哥伦比亚)
RBF-MGN:Solving spatiotemporal PDEs with Physics-informed Graph Neural Network
College of Aerospace Science and Engineering, National University of Defense Technology(国防科技大学航天科学与工程学院);National Innovation Institute of Defense Technology, Chinese Academy of Military Science(中国人民解放军军事科学院国防科技创新研究院)
Deep vs. Shallow: Benchmarking Physics-Informed Neural Architectures on the Biharmonic Equation
Department of Mechanical Engineering(机械工程系);Indian Institute of Technology, Madras(印度理工学院马德拉斯分校);Creatis Biomedical Imaging Laboratory(生物医学成像实验室);INSA-Lyon, France(法国里昂国立应用科学学院);Department of Data Science and AI(数据科学与人工智能系)