Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition

Abstract

Spectral decomposition of the Koopman operator is attracting attention as a tool for the analysis of nonlinear dynamical systems. Dynamic mode decomposition is a popular numerical algorithm for Koopman spectral analysis; however, we often need to prepare nonlinear observables manually according to the underlying dynamics, which is not always possible since we may not have any a priori knowledge about them. In this paper, we propose a fully data-driven method for Koopman spectral analysis based on the principle of learning Koopman invariant subspaces from observed data. To this end, we propose minimization of the residual sum of squares of linear least-squares regression to estimate a set of functions that transforms data into a form in which the linear regression fits well. We introduce an implementation with neural networks and evaluate performance empirically using nonlinear dynamical systems and applications.

18 pages, 7 figures, presented in NIPS 2017

References in corpus (3)

• Deep learning for universal linear embeddings of nonlinear dynamics
• Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
• Non-uniform state space reconstruction and coupling detection

Cited by in corpus (19)

• Deep learning for universal linear embeddings of nonlinear dynamics
• Data-driven discovery of coordinates and governing equations
• Discovering physical concepts with neural networks
• Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flows
• Transformers for Modeling Physical Systems
• Deep learning of dynamics and signal-noise decomposition with time-stepping constraints
• Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
• Uniformly Accurate Machine Learning Based Hydrodynamic Models for Kinetic Equations
• Characterizing Magnetized Plasmas with Dynamic Mode Decomposition
• Sparsity-promoting algorithms for the discovery of informative Koopman invariant subspaces
• Parameterized Neural Ordinary Differential Equations: Applications to Computational Physics Problems
• A data-driven Koopman model predictive control framework for nonlinear flows
• Data-driven Analysis for Understanding Team Sports Behaviors
• Efficient space-time reduced order model for linear dynamical systems in Python using less than 120 lines of code
• Koopman based data-driven predictive control
• Physically-interpretable classification of biological network dynamics for complex collective motions
• Two methods to approximate the Koopman operator with a reservoir computer
• Deep Adversarial Koopman Model for Reaction-Diffusion systems
• Operator Autoencoders: Learning Physical Operations on Encoded Molecular Graphs