System identification using autoregressive Bayesian neural networks with nonparametric noise models
From MaRDI portal
Abstract: System identification is of special interest in science and engineering. This article is concerned with a system identification problem arising in stochastic dynamic systems, where the aim is to estimate the parameters of a system along with its unknown noise processes. In particular, we propose a Bayesian nonparametric approach for system identification in discrete time nonlinear random dynamical systems assuming only the order of the Markov process is known. The proposed method replaces the assumption of Gaussian distributed error components with a highly flexible family of probability density functions based on Bayesian nonparametric priors. Additionally, the functional form of the system is estimated by leveraging Bayesian neural networks which also leads to flexible uncertainty quantification. Asymptotically on the number of hidden neurons, the proposed model converges to full nonparametric Bayesian regression model. A Gibbs sampler for posterior inference is proposed and its effectiveness is illustrated on simulated and real time series.
Cites work
- A Bayesian nonparametric approach to reconstruction and prediction of random dynamical systems
- A Bayesian nonparametric study of a dynamic nonlinear model
- A new Bayesian nonparametric mixture model
- A new look at the statistical model identification
- A nonparametric dependent process for Bayesian regression
- A survey of Bayesian predictive methods for model assessment, selection and comparison
- A widely applicable Bayesian information criterion
- Additive cubic spline regression with Dirichlet process mixture errors
- Applied nonlinear time series analysis. Applications in physics, physiology and finance.
- Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
- Bayes inference in regression models with ARMA\((p,q)\) errors
- Bayesian estimation of chaotic signals generated by piecewise-linear maps
- Bayesian learning for neural networks
- Bayesian reconstructions and predictions of nonlinear dynamical systems via the hybrid Monte Carlo scheme
- Chaos-based digital communication systems. Operating principles, analysis methods, and performance evaluation
- Dependent mixture models: clustering and borrowing information
- Dependent mixtures of geometric weights priors
- Dynamic systems identification with Gaussian processes
- Embedding and learning with signatures
- Finite mixture and Markov switching models.
- Forecasting nonlinear time series with a hybrid methodology
- Gaussian Process Latent Force Models for Learning and Stochastic Control of Physical Systems
- Gaussian processes for machine learning.
- Geometric stick-breaking processes for continuous-time Bayesian nonparametric modeling
- scientific article; zbMATH DE number 1573291 (Why is no real title available?)
- scientific article; zbMATH DE number 3742453 (Why is no real title available?)
- Joint reconstruction and prediction of random dynamical systems under borrowing of strength
- MCMC using Hamiltonian dynamics
- Neural networks for nonlinear dynamic system modelling and identification
- Nonlinear black-box modeling in system identification: A unified overview
- On the inferential implications of decreasing weight structures in mixture models
- Optimal tuning of the hybrid Monte Carlo algorithm
- Reconstructions and predictions of nonlinear dynamical systems: a hierarchical Bayesian approach
- Spike and slab variable selection: frequentist and Bayesian strategies
- System identification. An introduction.
- The horseshoe estimator for sparse signals
- Time series forecasting using a hybrid ARIMA and neural network model
- Time series modeling of neuroscience data
This page was built for publication: System identification using autoregressive Bayesian neural networks with nonparametric noise models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6135347)
