Ordinary and partial differential equations (ODEs/PDEs) play a paramount role in analyzing and simulating complex dynamic processes across all corners of science and engineering. In recent years machine learning tools are aspiring to introduce new effective ways of simulating PDEs, however existing approaches are not able to reliably return stable and accurate predictions across long temporal horizons. We aim to address this challenge by introducing an effective framework for learning infinite-dimensional operators that map random initial conditions to associated PDE solutions within a short time interval. Such latent operators can be parametrized by deep neural networks that are trained in an entirely self-supervised manner without requiring any paired input-output observations. Global long-time predictions across a range of initial conditions can be then obtained by iteratively evaluating the trained model using each prediction as the initial condition for the next evaluation step. This introduces a new approach to temporal domain decomposition that is shown to be effective in performing accurate long-time simulations for a wide range of parametric ODE and PDE systems, from wave propagation, to reaction-diffusion dynamics and stiff chemical kinetics, all at a fraction of the computational cost needed by classical numerical solvers.
Physics-Informed DeepONet Coupled with FEM for Convective Transport in Porous Media with Sharp Gaussian Sources
Department of Mathematics, Spelman College(斯帕尔曼学院数学系);Advanced Computing, Mathematics and Data Division, Pacific Northwest National Laboratory(太平洋西北国家实验室高级计算、数学与数据 division)
Improving physics-informed DeepONets with hard constraints
University of Hamburg(汉堡大学);Memorial University of Newfoundland(纽芬兰纪念大学);Silesian University in Opava(俄帕瓦斯莱西亚大学);Institute of Mathematics of NAS of Ukraine(乌克兰国家科学院数学研究所)
Physics-informed DeepONet with stiffness-based loss functions for structural response prediction
New York University Abu Dhabi(纽约大学阿布扎比分校);Nanjing University of Aeronautics and Astronautics(南京航空航天大学);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Mansoura University(曼苏拉大学)
PIP$^2$ Net: Physics-informed Partition Penalty Deep Operator Network
Hongjin Mi School of Mathematics Shanghai University of Finance and Economics(Hongjin Mi 数学学院 上海金融学院);Huiqiang Lun Faculty of Liberal Arts and Professional Studies York University(Huiqiang Lun 文科与专业学院 约克大学);Changhong Mou Department of Mathematics and Statistics Utah State University(Changhong Mou 数学与统计学系 Utah州立大学);Yeyu Zhang School of Mathematics Shanghai University of Finance and Economics(Yeyu Zhang 数学学院 上海金融学院)
Deep Operator Networks~(DeepONet)是一种完全不同的神经网络,我们训练它来近似非线性算子,包括参数偏微分方程(PDE)的求解算子。DeepONets已经显示出卓越的近似和泛化能力,即使在使用相对较小的数据集进行训练时也是如此。然而,当训练数据被噪声污染时,DeepONets的性能会恶化,这种情况在实践中经常发生。为了使DeepONets能够使用噪声数据进行训练,我们建议使用复制交换Langevin扩散的贝叶斯框架。这样的框架使用两个粒子,一个用于探索,另一个用于利用DeepONets的损失函数。我们表明,所提出的框架的探索和利用能力使(1)在嘈杂的场景中提高DeepONets的训练收敛性,以及(2)为参数PDE的预测解附加不确定性估计。此外,我们还发现,与使用最先进的基于梯度的优化算法(例如Adam)训练的香草DeepONets相比,复制交换Langeving扩散(显着)还提高了DeepONet在嘈杂场景中的平均预测准确性。为了降低副本的潜在高计算成本,在这项工作中,我们提出了一个用于复制交换Langevin扩散的加速训练框架,该框架利用DeepONets的神经网络架构将其计算成本降低了25%,而不会影响所提出的框架的性能。最后,我们使用一系列的实验四个参数偏微分方程的问题,说明了所提出的贝叶斯框架的有效性。
fPINN-DeepONet: A Physics-Informed Operator Learning Framework for Multi-term Time-fractional Mixed Diffusion-wave Equations
Department of Mathematics, Purdue University West Lafayette, IN 47907, USA(普渡大学数学系);Department of Mathematics, Purdue University School of Mechanical Engineering, Purdue University West Lafayette, IN 47907, USA(普渡大学机械工程学院数学系)
Physics-Informed Time-Integrated DeepONet: Temporal Tangent Space Operator Learning for High-Accuracy Inference
Institute of Structural Mechanics and Dynamics in Aerospace Engineering, University of Stuttgart(航空航天工程力学与动力学研究所,斯图加特大学);Experimental Hepatobiliary Surgery Group, Department of Hepatobiliary Surgery and Visceral Transplantation, University of Leipzig Medical Center(肝胆外科与 visceral 移植部门实验肝胆外科小组,莱比锡大学医学中心);Department of Civil and Systems Engineering, Johns Hopkins University(土木与系统工程系,约翰霍普金斯大学)
An Advanced Physics-Informed Neural Operator for Comprehensive Design Optimization of Highly-Nonlinear Systems: An Aerospace Composites Processing Case Study
An explainable operator approximation framework under the guideline of Green's function
State Key Laboratory of Turbulence and Complex Systems, College of Engineering, Peking University(湍流与复杂系统国家重点实验室,工程学院,北京大学);Ningbo Institute of Digital Twin, Eastern Institute of Technology(宁波数字孪生研究所,东部技术研究所);Zhejiang Key Laboratory of Industrial Intelligence and Digital Twin, Eastern Institute of Technology(浙江省工业智能与数字孪生重点实验室,东部技术研究所)
Multifidelity Deep Operator Networks For Data-Driven and Physics-Informed Problems
Pacific Northwest National Laboratory(太平洋西北国家实验室);Sandia National Laboratories(桑迪亚国家实验室);Division of Applied Mathematics and School of Engineering, Brown University(布朗大学应用数学系与工程学院)
Physics-informed deep operator network for traffic state estimation
The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Tongji University, Shanghai 201804, China(交通工程重点实验室,教育部,同济大学,上海201804,中国);College of Transportation Engineering, Tongji University, Shanghai 201804, China(交通工程学院,同济大学,上海201804,中国)
Learning Physics-Informed Surrogate Model of Linear Elastic Displacement Fields from Geometry
PROMES-CNRS, Université Perpignan Via Domitia (UPVD)(法国国家科学研究中心佩皮尼昂大学多姆itia分校能源与过程系统优化实验室);LAAS-CNRS, Université de Toulouse (UT)(法国国家科学研究中心图卢兹大学航空航天系统实验室)
On the training of physics-informed neural operators for solving parametric partial differential equations
College of Civil Engineering, Tongji University(同济大学土木工程学院);Department of Engineering Science, University of Oxford(牛津大学工程科学系);Institute for Foundations of Data Science, Yale University(耶鲁大学数据科学基础研究所)
Disentangling Attention in Deep Operator Learning: A Controlled Study of Data-Driven and Physics-Informed Architectures
National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校国家超级计算应用中心);Department of Mechanical Engineering, Stanford University(斯坦福大学机械工程系)
LFR-PINO: A Layered Fourier Reduced Physics-Informed Neural Operator for Parametric PDEs
School of Aeronautics and Astronautics, Shanghai Jiao Tong University, Shanghai, China(上海交通大学航空宇航科学与技术学院);School of Aeronautics and Astronautics, Zhejiang University, Hangzhou, China(浙江大学航空宇航科学与技术学院);Key Laboratory for Satellite Digitalization Technology, Innovation Academy for Microsatellites of CAS, Shanghai, China(中国科学院卫星数字化技术重点实验室)
VAE-DNN: Energy-Efficient Trainable-by-Parts Surrogate Model For Parametric Partial Differential Equations
School of Water Conservancy and Transportation(水利与交通学院);Zhengzhou University(郑州大学);Department of Civil and Environmental Engineering(土木与环境工程系);University of Illinois Urbana Champaign(伊利诺伊大学厄巴纳-香槟分校)
Muon with Spectral Guidance: Efficient Optimization for Scientific Machine Learning
Elmore Family School of Electrical and Computer Engineering(埃洛姆家族电气与计算机工程学院);Purdue University(普渡大学);Department of Mathematics(数学系);Department of Mathematics & School of Mechanical Engineering(数学系与机械工程学院)
Physics-informed neural operator for parametric phase-field modelling of interfacial degradation and microstructural evolution
Tongji University, College of Civil Engineering(同济大学,土木工程学院);Tongji University, State Key Laboratory of Disaster Reduction in Civil Engineering(同济大学,灾害防治工程重点实验室)
Crack Path Prediction with Operator Learning using Discrete Particle System data Generation
Division of Applied Mathematics, Brown University, USA(布朗大学应用数学系);School of Mechanical Sciences, Indian Institute of Technology Bhubaneswar, India(印度理工学院布巴内斯瓦尔分校机械科学学院);Department of Mechanical Engineering, Clemson University, USA(克莱姆斯大学机械工程系)