On the Role of Fixed Points of Dynamical Systems in Training Physics-Informed Neural Networks
Know-Center GmbH – Research Center for Data-Driven Business & Big Data Analytics(知识中心有限公司——数据驱动业务与大数据分析研究中心);LEC GmbH – Large Engines Competence Center(大型发动机竞争力中心有限公司)
Franz M. Rohrhofer, Stefan Posch, Clemens Gößnitzer, Bernhard C. Geiger
This paper empirically studies commonly observed training difficulties of Physics-Informed Neural Networks (PINNs) on dynamical systems. Our results indicate that fixed points which are inherent to these systems play a key role in the optimization of the in PINNs embedded physics loss function. We observe that the loss landscape exhibits local optima that are shaped by the presence of fixed points. We find that these local optima contribute to the complexity of the physics loss optimization which can explain common training difficulties and resulting nonphysical predictions. Under certain settings, e.g., initial conditions close to fixed points or long simulations times, we show that those optima can even become better than that of the desired solution.
Physics Informed Kolmogorov-Arnold Neural Networks for Dynamical Analysis via Efficent-KAN and WAV-KAN
Indian Institute of Technology, Delhi(印度理工学院德里分校);Indian Institute of Technology, Dhanbad(印度理工学院丹巴德分校);Sun Moon University(日月大学);Amrita Vishwa Vidyapeetham(阿米塔大学);United States DEVCOM Army Research Laboratory(美国DEVCOM陆军研究实验室);University of Massachusetts at Boston(马萨诸塞大学波士顿分校);Tulane University(杜兰大学)
Residual-loss Anomaly Analysis of Physics-Informed Neural Networks: An Inverse Method for Change-point Detection in Nonlinear Dynamical Systems with Regime Switching
School of Mathematics and Statistics(数学与统计学学院);Huazhong University of Science and Technology(华中科技大学);Northwestern Polytechnical University(西北工业大学)
Physics-Informed Neural Network Frameworks for the Analysis of Engineering and Biological Dynamical Systems Governed by Ordinary Differential Equations
Siena Heights University(锡耶纳高地大学);Wayne State University(韦恩州立大学)
PINN-Obs: Physics-Informed Neural Network-Based Observer for Nonlinear Dynamical Systems
TICLab, International University of Rabat, Rabat 11100, Morocco(拉巴特国际大学TIC实验室);CNRS–CRAN–7039, University of Lorraine, France(法国洛林大学CNRS–CRAN–7039机构);Faculty of Engineering, University of Leeds, LS2 9JT, United Kingdom(利兹大学工程学院);INRIA Nancy—LARSEN, France(法国INRIA南锡—LARSEN机构)
Deep Neural Networks as Discrete Dynamical Systems: Implications for Physics-Informed Learning
Engineering Mechanics Unit, Jawaharlal Nehru Centre for Advanced Scientific Research(纳拉扬·德赛高级科学研究中心工程力学单元);Italian Institute of Technology(意大利理工学院);University of Roma Tre(罗马三大学);Physics Department, Harvard University(哈佛大学物理系);Cornell University(康奈尔大学)
Munich Data Science Institute(慕尼黑数据科学研究所);School of Computation, Information and Technology, Technical University of Munich(技术大学慕尼黑计算、信息与技术学院);Interdisciplinary Center for Scientific Computing (IWR)(跨学科科学计算中心);Department of Mathematics and Computer Science, Heidelberg University(海德堡大学数学与计算机科学系)
A Rapid Physics-Informed Machine Learning Framework Based on Extreme Learning Machine for Inverse Stefan Problems
School of Qilu Transportation, Shandong University(山东大学齐鲁交通学院);School of Mechanical Engineering, Qilu University of Technology (Shandong Academy of Sciences)(齐鲁工业大学(山东省科学院)机械工程学院)
Adapting Physics-Informed Neural Networks to Improve ODE Optimization in Mosquito Population Dynamics
School of Computing, Dublin City University(都柏林城市大学计算学院);Faculty of Agriculture, University of Novi Sad(诺威萨大学农业学院);Avia-GIS NV(Avia-GIS NV公司);University of Science, Ho Chi Minh City(胡志明市科学大学);Insight Centre for Data Analytics, Dublin City University(都柏林城市大学洞察数据分析中心)
Invariant Manifolds of Discrete-time Dynamical Systems with Nonlinear Exosystems via Hybrid Physics-Informed Neural Networks
Modelling Engineering Risk & Complexity, Scuola Superiore Meridionale(工程风险与复杂性建模,南方高级学校);Dept. of Chemical Engineering, Worcester Polytechnic Institute(化学工程系,沃思希尔理工大学);Dept. of Chemical & Biomolecular Engineering, Johns Hopkins University(化学与生物分子工程系,约翰霍普金斯大学);Dept. of Applied Mathematics and Statistics, Johns Hopkins University(应用数学与统计学系,约翰霍普金斯大学);Medical School, Department of Urology, Johns Hopkins University(医学院,泌尿科,约翰霍普金斯大学);Institute of Science and Technology for Energy and Sustainable Mobility, Consiglio Nazionale delle Ricerche(能源与可持续交通科学技术研究所,国家研究委员会)
PIDM-DP: Physics-Informed Diffusion with Dormand-Prince Integration for Chaotic System Identification and State Reconstruction across Multiple Dynamical Regimes
ODIN: ODE-Informed Regression for Parameter and State Inference in Time-Continuous Dynamical Systems
ETH Zürich(苏黎世联邦理工学院);Max Planck ETH Center for Learning Systems(马克斯·普朗克-苏黎世联邦理工学院学习系统中心);University of Oxford(牛津大学);Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所)
Physics-Informed Machine Learning for Modeling and Control of Dynamical Systems
Northern Arizona University(北亚利桑那大学);Pacific Northwest National Laboratory(太平洋西北国家实验室);EPFL(洛桑联邦理工学院);The University of Texas at Austin(德克萨斯大学奥斯汀分校);Siemens Corporation(西门子公司);Mitsubishi Electric Research Laboratories(三菱电机研究实验室);The Ohio State University(俄亥俄州立大学);ETH Zurich(苏黎世联邦理工学院)
Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systems
Massachusetts Institute of Technology(麻省理工学院);University of California San Diego(加州大学圣地亚哥分校);New York University(纽约大学);University of Texas at Austin(德克萨斯大学奥斯汀分校)