Physics-Informed Probabilistic Learning of Linear Embeddings of Non-linear Dynamics With Guaranteed Stability

Abstract

The Koopman operator has emerged as a powerful tool for the analysis of nonlinear dynamical systems as it provides coordinate transformations to globally linearize the dynamics. While recent deep learning approaches have been useful in extracting the Koopman operator from a data-driven perspective, several challenges remain. In this work, we formalize the problem of learning the continuous-time Koopman operator with deep neural networks in a measure-theoretic framework. Our approach induces two types of models: differential and recurrent form, the choice of which depends on the availability of the governing equations and data. We then enforce a structural parameterization that renders the realization of the Koopman operator provably stable. A new autoencoder architecture is constructed, such that only the residual of the dynamic mode decomposition is learned. Finally, we employ mean-field variational inference (MFVI) on the aforementioned framework in a hierarchical Bayesian setting to quantify uncertainties in the characterization and prediction of the dynamics of observables. The framework is evaluated on a simple polynomial system, the Duffing oscillator, and an unstable cylinder wake flow with noisy measurements.

arXiv comment31 pages

References in corpus (31)

• Adam: A Method for Stochastic Optimization
• TensorFlow: A system for large-scale machine learning
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Variational Inference: A Review for Statisticians
• A Data-Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition
• Deep learning for universal linear embeddings of nonlinear dynamics
• Weight Uncertainty in Neural Networks
• Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Searching for Activation Functions
• Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
• Applied Koopmanism
• Koopman invariant subspaces and finite linear representations of nonlinear dynamical systems for control
• Ergodic theory, Dynamic Mode Decomposition and Computation of Spectral Properties of the Koopman operator
• Extended dynamic mode decomposition with dictionary learning: a data-driven adaptive spectral decomposition of the Koopman operator
• Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
• Stable Architectures for Deep Neural Networks
• On Convergence of Extended Dynamic Mode Decomposition to the Koopman Operator
• Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm
• Edward: A library for probabilistic modeling, inference, and criticism
• Study of dynamics in post-transient flows using Koopman mode decomposition
• Geometry of the ergodic quotient reveals coherent structures in flows
• AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks
• A Kernel-Based Approach to Data-Driven Koopman Spectral Analysis
• A Priori Estimation Of Memory Effects In Coarse-Grained Nonlinear Systems Using The Mori-Zwanzig Formalism
• A Dynamic Subgrid Scale Model for Large Eddy Simulations Based on the Mori-Zwanzig Formalism
• Non-Markovian Closure Models for Large Eddy Simulations using the Mori-Zwanzig Formalism
• Message Passing Stein Variational Gradient Descent
• A data-driven Koopman model predictive control framework for nonlinear flows
• Stein Variational Gradient Descent as Moment Matching
• Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems

Cited by in corpus (41)

• Fourier Neural Operator for Parametric Partial Differential Equations
• Multi-level Convolutional Autoencoder Networks for Parametric Prediction of Spatio-temporal Dynamics
• On closures for reduced order models $-$ A spectrum of first-principle to machine-learned avenues
• Physics guided machine learning using simplified theories
• Learning Deep Implicit Fourier Neural Operators (IFNOs) with Applications to Heterogeneous Material Modeling
• Assessment of unsteady flow predictions using hybrid deep learning based reduced order models
• Data-driven recovery of hidden physics in reduced order modeling of fluid flows
• Physics-constrained, low-dimensional models for MHD: First-principles and data-driven approaches
• On the Structure of Time-delay Embedding in Linear Models of Non-linear Dynamical Systems
• Attention-Enhanced Neural Network Models for Turbulence Simulation
• Sparsity-promoting algorithms for the discovery of informative Koopman invariant subspaces
• Algorithmically-Consistent Deep Learning Frameworks for Structural Topology Optimization
• Parametric Dynamic Mode Decomposition for Reduced Order Modeling
• Nonlinear proper orthogonal decomposition for convection-dominated flows
• From Fourier to Koopman: Spectral Methods for Long-term Time Series Prediction
• Nonlocal Kernel Network (NKN): a Stable and Resolution-Independent Deep Neural Network
• Deep Learning Models for Global Coordinate Transformations that Linearize PDEs
• Incorporating physical constraints in a deep probabilistic machine learning framework for coarse-graining dynamical systems
• Forecasting Sequential Data using Consistent Koopman Autoencoders
• Identification of MIMO Wiener-type Koopman Models for Data-Driven Model Reduction using Deep Learning
• Koopman operator learning using invertible neural networks
• Data-driven Modeling of Rotating Detonation Waves
• Structure Probing Neural Network Deflation
• Two methods to approximate the Koopman operator with a reservoir computer
• CD-ROM: Complemented Deep-Reduced Order Model
• CGKN: A Deep Learning Framework for Modeling Complex Dynamical Systems and Efficient Data Assimilation
• Koopman Operator Dynamical Models: Learning, Analysis and Control
• Modeling Partially Observed Nonlinear Dynamical Systems and Efficient Data Assimilation via Discrete-Time Conditional Gaussian Koopman Network
• Promoting global stability in data-driven models of quadratic nonlinear dynamics
• Optimal Dirichlet Boundary Control by Fourier Neural Operators Applied to Nonlinear Optics
• Learning phase field mean curvature flows with neural networks
• Noisy Recurrent Neural Networks
• Stochastic Adversarial Koopman Model for Dynamical Systems
• Quantum System Compression: A Hamiltonian Guided Walk Through Hilbert Space
• Forecasting the evolution of three-dimensional turbulent recirculating flows from sparse sensor data
• Deep Adversarial Koopman Model for Reaction-Diffusion systems
• Physics-aware, probabilistic model order reduction with guaranteed stability
• Nonintrusive reduced order modeling of convective Boussinesq flows
• A purely data-driven framework for prediction, optimization, and control of networked processes: application to networked SIS epidemic model
• An evolve-then-correct reduced order model for hidden fluid dynamics
• Learning Stable Koopman Embeddings