Extended dynamic mode decomposition with dictionary learning using neural ordinary differential equations

被引:2
|
作者
Terao, Hiroaki [1 ]
Shirasaka, Sho [1 ]
Suzuki, Hideyuki [1 ]
机构
[1] Osaka Univ, Grad Sch Informat Sci & Technol, 1-5 Yamadaoka, Suita, Osaka 5650871, Japan
来源
关键词
time series analysis; Koopman operator; extended dynamic mode decomposition; machine learning; dictionary learning; neural ordinary differential equations; SYSTEMS;
D O I
10.1587/nolta.12.626
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Nonlinear phenomena can be analyzed via linear techniques using operator-theoretic approaches. Data-driven method called the extended dynamic mode decomposition (EDMD) and its variants, which approximate the Koopman operator associated with the nonlinear phenomena, have been rapidly developing by incorporating machine learning methods. Neural ordinary differential equations (NODEs), which are a neural network equipped with a continuum of layers, and have high parameter and memory efficiencies, have been proposed. In this paper, we propose an algorithm to perform EDMD using NODEs. NODEs are used to find a parameter-efficient dictionary which provides a good finite-dimensional approximation of the Koopman operator. We show the superiority of the parameter efficiency of the proposed method through numerical experiments.
引用
收藏
页码:626 / 638
页数:13
相关论文
共 50 条
  • [1] Extended Dynamic Mode Decomposition with Invertible Dictionary Learning
    Jin, Yuhong
    Hou, Lei
    Zhong, Shun
    [J]. NEURAL NETWORKS, 2024, 173
  • [2] Extended dynamic mode decomposition with dictionary learning: A data-driven adaptive spectral decomposition of the Koopman operator
    Li, Qianxiao
    Dietrich, Felix
    Bollt, Erik M.
    Kevrekidis, Ioannis G.
    [J]. CHAOS, 2017, 27 (10)
  • [3] Decomposition of ordinary differential equations
    Schwarz, Fritz
    [J]. BULLETIN OF MATHEMATICAL SCIENCES, 2017, 7 (03) : 575 - 613
  • [4] Neural Ordinary Differential Equations
    Chen, Ricky T. Q.
    Rubanova, Yulia
    Bettencourt, Jesse
    Duvenaud, David
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 31 (NIPS 2018), 2018, 31
  • [5] Semantic Segmentation Using Neural Ordinary Differential Equations
    Khoshsirat, Seyedalireza
    Kambhamettu, Chandra
    [J]. ADVANCES IN VISUAL COMPUTING, ISVC 2022, PT I, 2022, 13598 : 284 - 295
  • [6] Solving Ordinary Differential Equations Using Neural Networks
    Sen Tan, Lee
    Zainuddin, Zarita
    Ong, Pauline
    [J]. PROCEEDING OF THE 25TH NATIONAL SYMPOSIUM ON MATHEMATICAL SCIENCES (SKSM25): MATHEMATICAL SCIENCES AS THE CORE OF INTELLECTUAL EXCELLENCE, 2018, 1974
  • [7] A Forward Learning Algorithm for Neural Memory Ordinary Differential Equations
    Xu, Xiuyuan
    Luo, Haiying
    Yi, Zhang
    Zhang, Haixian
    [J]. INTERNATIONAL JOURNAL OF NEURAL SYSTEMS, 2024, 34 (09)
  • [8] Learning quantum dynamics with latent neural ordinary differential equations
    Choi, Matthew
    Flam-Shepherd, Daniel
    Kyaw, Thi Ha
    Aspuru-Guzik, Alan
    [J]. PHYSICAL REVIEW A, 2022, 105 (04)
  • [9] Inverter Dynamic Response Simulation via Neural Ordinary Differential Equations
    Liu, Yi
    Huang, Qiuhua
    [J]. 2024 INTERNATIONAL WORKSHOP ON ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING FOR ENERGY TRANSFORMATION, AIE 2024, 2024,
  • [10] Dynamic mode decomposition via dictionary learning for foreground modeling in videos
    Ul Haq, Israr
    Fujii, Keisuke
    Kawahara, Yoshinobu
    [J]. COMPUTER VISION AND IMAGE UNDERSTANDING, 2020, 199