A Generalizable Physics-informed Learning Framework for Risk Probability Estimation

被引:0
|
作者
Wang, Zhuoyuan [1 ]
Nakahira, Yorie [1 ]
机构
[1] Carnegie Mellon Univ, Elect & Comp Engn Dept, Pittsburgh, PA 15213 USA
关键词
Stochastic safe control; physics-informed learning; risk probability estimation; SAFE EXPLORATION; REACHABILITY;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Accurate estimates of long-term risk probabilities and their gradients are critical for many stochastic safe control methods. However, computing such risk probabilities in real-time and in unseen or changing environments is challenging. Monte Carlo (MC) methods cannot accurately evaluate the probabilities and their gradients as an infinitesimal devisor can amplify the sampling noise. In this paper, we develop an efficient method to evaluate the probabilities of long-term risk and their gradients. The proposed method exploits the fact that long-term risk probability satisfies certain partial differential equations (PDEs), which characterize the neighboring relations between the probabilities, to integrate MC methods and physics-informed neural networks. We provide theoretical guarantees of the estimation error given certain choices of training configurations. Numerical results show the proposed method has better sample efficiency, generalizes well to unseen regions, and can adapt to systems with changing parameters. The proposed method can also accurately estimate the gradients of risk probabilities, which enables first- and second-order techniques on risk probabilities to be used for learning and control.
引用
收藏
页数:13
相关论文
共 50 条
  • [1] Physics-Informed Spatiotemporal Learning Framework for Urban Traffic State Estimation
    Shi, Zeyu
    Chen, Yangzhou
    Liu, Jichao
    Fan, Dechao
    Liang, Chaoqiang
    JOURNAL OF TRANSPORTATION ENGINEERING PART A-SYSTEMS, 2023, 149 (07)
  • [2] Generalizable synthetic MRI with physics-informed convolutional networks
    Jacobs, Luuk
    Mandija, Stefano
    Liu, Hongyan
    van den Berg, Cornelis A. T.
    Sbrizzi, Alessandro
    Maspero, Matteo
    MEDICAL PHYSICS, 2024, 51 (05) : 3348 - 3359
  • [3] Physics-informed machine learning
    George Em Karniadakis
    Ioannis G. Kevrekidis
    Lu Lu
    Paris Perdikaris
    Sifan Wang
    Liu Yang
    Nature Reviews Physics, 2021, 3 : 422 - 440
  • [4] Physics-informed machine learning
    Karniadakis, George Em
    Kevrekidis, Ioannis G.
    Lu, Lu
    Perdikaris, Paris
    Wang, Sifan
    Yang, Liu
    NATURE REVIEWS PHYSICS, 2021, 3 (06) : 422 - 440
  • [5] A Framework for Physics-Informed Deep Learning Over Freeform Domains
    Mezzadri, Francesco
    Gasick, Joshua
    Qian, Xiaoping
    COMPUTER-AIDED DESIGN, 2023, 160
  • [6] Physics-Informed Graph Learning
    Peng, Ciyuan
    Xia, Feng
    Saikrishna, Vidya
    Liu, Huan
    2022 IEEE INTERNATIONAL CONFERENCE ON DATA MINING WORKSHOPS, ICDMW, 2022, : 732 - 739
  • [7] PHYSICS-INFORMED GENERALIZABLE WIRELESS CHANNEL MODELING WITH SEGMENTATION AND DEEP LEARNING: FUNDAMENTALS, METHODOLOGIES, AND CHALLENGES
    Zhu, Ethan
    Sun, Haijian
    Ji, Mingyue
    IEEE WIRELESS COMMUNICATIONS, 2024, : 170 - 177
  • [8] Separable physics-informed DeepONet: Breaking the curse of dimensionality in physics-informed machine learning
    Mandl, Luis
    Goswami, Somdatta
    Lambers, Lena
    Ricken, Tim
    Computer Methods in Applied Mechanics and Engineering, 2025, 434
  • [9] Sensor Fault Diagnostics Using Physics-Informed Transfer Learning Framework
    Guc, Furkan
    Chen, Yangquan
    SENSORS, 2022, 22 (08)
  • [10] Parameter Estimation of Power Electronic Converters With Physics-Informed Machine Learning
    Zhao, Shuai
    Peng, Yingzhou
    Zhang, Yi
    Wang, Huai
    IEEE TRANSACTIONS ON POWER ELECTRONICS, 2022, 37 (10) : 11567 - 11578