Error Bounds for Kernel-Based Linear System Identification with Unknown Hyperparameters
Preprint
Abstract
The kernel-based method has been successfully applied in linear system identification using stable kernel designs. From a Gaussian process perspective, it automatically provides probabilistic error bounds for the identified models from the posterior covariance, which are useful in robust and stochastic control. However, the error bounds require knowledge of the true hyperparameters in the kernel design and are demonstrated to be inaccurate with estimated hyperparameters for lightly damped systems or in the presence of high noise. In this work, we provide reliable quantification of the estimation error when the hyperparameters are unknown. The bounds are obtained by first constructing a high-probability set for the true hyperparameters from the marginal likelihood function and then finding the worst-case posterior covariance within the set. The proposed bound is proven to contain the true model with a high probability and its validity is verified in numerical simulation.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Deterministic error bounds for kernel-based learning techniques under bounded noiseAutomatica, 2021
- Sample Complexity and Minimax Properties of Exponentially Stable Regularized EstimatorsIEEE Transactions on Automatic Control, 2021
- System Identification: A Machine Learning PerspectiveAnnual Review of Control, Robotics, and Autonomous Systems, 2019
- A shift in paradigm for system identificationInternational Journal of Control, 2019
- On kernel design for regularized LTI system identificationAutomatica, 2018
- Regularized linear system identification using atomic, nuclear and kernel-based norms: The role of the stability constraintAutomatica, 2016
- On the estimation of transfer functions, regularizations and Gaussian processes—RevisitedAutomatica, 2012
- Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit SettingIEEE Transactions on Information Theory, 2012
- A new kernel-based approach for linear system identificationAutomatica, 2010