Recurrent Deep Kernel Learning of Dynamical Systems
Not on the map. Why?
Abstract (arXiv)
Digital twins require computationally-efficient reduced-order models (ROMs) that can accurately describe complex dynamics of physical assets. However, constructing ROMs from noisy high-dimensional data is challenging. In this work, we propose a data-driven, non-intrusive method that utilizes stochastic variational deep kernel learning (SVDKL) to discover low-dimensional latent spaces from data and a recurrent version of SVDKL for representing and predicting the evolution of latent dynamics. The proposed method is demonstrated with two challenging examples -- a double pendulum and a reaction-diffusion system. Results show that our framework is capable of (i) denoising and reconstructing measurements, (ii) learning compact representations of system states, (iii) predicting system evolution in low-dimensional latent spaces, and (iv) quantifying modeling uncertainties.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 23 of its 59 references that are in the map. First posted on arXiv 30 May 2024.
Builds on
23 of 59 references.
References · 23
- 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,598
- 2016Variational Inference: A Review for Statisticianscited by 1,466
- 2018Learning Latent Dynamics for Planning from Pixelscited by 1,221
- 2015Deep Kernel Learningcited by 440
- 2017Linearly-Recurrent Autoencoder Networks for Learning Dynamicscited by 142
- 2020Bayesian Neural Networks: An Introduction and Surveycited by 130
- 2020A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized PDEscited by 105
- 2019Lift & Learn: Physics-informed machine learning for large-scale nonlinear dynamical systemscited by 96
- 2016Stochastic Variational Deep Kernel Learningcited by 92
- 2021POD-DL-ROM: enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decompositioncited by 79
- 2022Discovering Governing Equations from Partial Measurements with Deep Delay Autoencoderscited by 58
- 2018Kernel Flows: from learning kernels from data into the abysscited by 49
11 more
- 2022Reduced order modeling of parametrized systems through autoencoders and SINDy approach: continuation of periodic solutionscited by 29
- 2022Bayesian autoencoders for data-driven discovery of coordinates, governing equations and fundamental constantscited by 24
- 2014Data-Driven Model Reduction for the Bayesian Solution of Inverse Problemscited by 20
- 2022Bayesian operator inference for data-driven reduced-order modelingcited by 19
- 2023Multi-fidelity reduced-order surrogate modelingcited by 18
- 2021Data-driven geophysical forecasting: Simple, low-cost, and accurate baselines with kernel methodscited by 15
- 2022One-Shot Learning of Stochastic Differential Equations with Data Adapted Kernelscited by 15
- 2023Learning Dynamical Systems from Data: A Simple Cross-Validation Perspective, Part V: Sparse Kernel Flows for 132 Chaotic Dynamical Systemscited by 15
- 2014Preserving Lagrangian structure in nonlinear model reduction with application to structural dynamicscited by 10
- 2021Model Order Reduction based on Runge-Kutta Neural Networkcited by 9
- 2022Deep Kernel Learning of Dynamical Models from High-Dimensional Noisy Datacited by 6
Built on by
Citing papers per year: 2025: 1.
1 paper.