How entropic regression beats the outliers problem in nonlinear system identification
From MaRDI portal
Publication:5218142
DOI10.1063/1.5133386zbMATH Open1433.93032arXiv1905.08061OpenAlexW2997092692WikidataQ89512435 ScholiaQ89512435MaRDI QIDQ5218142FDOQ5218142
Authors: Abd AlRahman R. AlMomani, Jie Sun, Erik Bollt 
Publication date: 28 February 2020
Published in: Chaos: An Interdisciplinary Journal of Nonlinear Science (Search for Journal in Brave)
Abstract: In this work, we developed a nonlinear System Identification (SID) method that we called Entropic Regression. Our method adopts an information-theoretic measure for the data-driven discovery of the underlying dynamics. Our method shows robustness toward noise and outliers and it outperforms many of the current state-of-the-art methods. Moreover, the method of Entropic Regression overcomes many of the major limitations of the current methods such as sloppy parameters, diverse scale, and SID in high dimensional systems such as complex networks. The use of information-theoretic measures in entropic regression poses unique advantages, due to the Asymptotic Equipartition Property (AEP) of probability distributions, that outliers and other low-occurrence events are conveniently and intrinsically de-emphasized as not-typical, by definition. We provide a numerical comparison with the current state-of-the-art methods in sparse regression, and we apply the methods to different chaotic systems such as the Lorenz System, the Kuramoto-Sivashinsky equations, and the Double Well Potential.
Full work available at URL: https://arxiv.org/abs/1905.08061
Recommendations
- Sparse identification of nonlinear dynamical systems via reweighted \(\ell_1\)-regularized least squares
- An entropy-like estimator for robust parameter identification
- An extended orthogonal forward regression algorithm for system identification using entropy
- Regularized LTI system identification in the presence of outliers: a variational EM approach
- Simultaneous Identification and Denoising of Dynamical Systems
Measures of information, entropy (94A17) System identification (93B30) Nonlinear systems in control theory (93C10)
Cites Work
- A new look at the statistical model identification
- Title not available (Why is that?)
- Title not available (Why is that?)
- Stable signal recovery from incomplete and inaccurate measurements
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
- Deterministic Nonperiodic Flow
- Compressed sensing
- The Kuramoto-Sivashinsky equation: a bridge between PDE's and dynamical systems
- Infinite-dimensional dynamical systems. An introduction to dissipative parabolic PDEs and the theory of global attractors
- Nonlinear analysis of hydrodynamic instability in laminar flames—I. Derivation of basic equations
- Approximate inertial manifolds for the Kuramoto-Sivashinsky equation: Analysis and computations
- Spatiotemporal chaos in terms of unstable recurrent patterns
- Orthogonal parameter estimation algorithm for non-linear stochastic systems
- An extended orthogonal forward regression algorithm for system identification using entropy
- Orthogonal least squares methods and their application to non-linear system identification
- Adaptive algorithms for sparse system identification
- Accurate computations on inertial manifolds
- Data based identification and prediction of nonlinear and complex dynamical systems
- Causal network inference by optimal causation entropy
- Modeling and nonlinear parameter estimation with Kronecker product representation for coupled oscillators and spatiotemporal systems
- Slow manifolds of some chaotic systems with applications to laser systems
- Maximum entropy vector kernels for MIMO system identification
Cited In (15)
- Adaptive integral alternating minimization method for robust learning of nonlinear dynamical systems from highly corrupted data
- Regularized least absolute deviation-based sparse identification of dynamical systems
- Entropic regression with neurologically motivated applications
- Knowledge-based learning of nonlinear dynamics and chaos
- Learning chaotic systems from noisy data via multi-step optimization and adaptive training
- A causation-based computationally efficient strategy for deploying Lagrangian drifters to improve real-time state estimation
- CGNSDE: conditional Gaussian neural stochastic differential equation for modeling complex systems and data assimilation
- SubTSBR to tackle high noise and outliers for data-driven discovery of differential equations
- The structure of reconstructed flows in latent spaces
- Sparse methods for automatic relevance determination
- Bridging the gap: machine learning to resolve improperly modeled dynamics
- Finding nonlinear system equations and complex network structures from data: a sparse optimization approach
- Sparse identification of dynamical systems by reweighted \(l_1\)-regularized least absolute deviation regression
- Symbolic regression via neural networks
- Model selection of chaotic systems from data with hidden variables using sparse data assimilation
Uses Software
This page was built for publication: How entropic regression beats the outliers problem in nonlinear system identification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5218142)
