Convergence rate determination for gradient-based adaptive estimators

DOI10.1016/0005-1098(86)90079-8MaRDI QIDQ1084367zbMATH OpenOpenAlexFDO

Authors Brian D. O. Anderson, Robert R. Bitmead, Tung-Sang Ng

Publication date 1986

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0005-1098(86)90079-8

zbMATH Keywords

exponential convergence rategradient-based adaptive parameter estimatorslower bound

Mathematics Subject Classification ID

Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Linear systems in control theory (93C05) Adaptive control/observation systems (93C40) Estimation and detection in stochastic control theory (93E10) Identification in stochastic control theory (93E12)

Review text

The paper presents a method for deriving a lower bound on the exponential convergence rate of the homogeneous part of gradient-based adaptive parameter estimators. The convergence rate depends on the minimum eigenvalue of the covariance matrix of the given random process. A comparison is drawn between adaptive estimators and the experimental design criteria for nonadaptive system identification.

Recommendations

• Rate of Convergence in Recursive Parameter Estimation procedures
• Exponential convergence of the Kalman filter based parameter estimation algorithm
• Convergence analysis of adaptive linear estimation for dependent stationary processes
• scientific article; zbMATH DE number 647688
• Convergence rate of least-squares identification and adaptive control for stochastic systems†

Cites work

• A measure of the tracking capability of recursive stochastic algorithms with constant gains
• An approach to nonlinear programming
• Asymptotic behavior of an adaptive estimation algorithm with application to M-dependent data
• Asymptotic Properties of Stochastic Approximations with Constant Coefficients
• Convergence in distribution of LMS-type adaptive parameter estimates
• Convergence properties of LMS adaptive estimators with unbounded dependent inputs
• Digital adaptive filters: Conditions for convergence, rates of convergence, effects of noise and errors arising from the implementation
• Dynamic system identification. Experiment design and data analysis
• scientific article; zbMATH DE number 3817512 (Why is no real title available?)
• scientific article; zbMATH DE number 3688420 (Why is no real title available?)
• scientific article; zbMATH DE number 3349105 (Why is no real title available?)
• scientific article; zbMATH DE number 3406941 (Why is no real title available?)
• Optimal Input Design for Dynamic System Parameter Estimation: The D_s-Optimality Case
• Performance of adaptive estimation algorithms in dependent random environments
• Second-order convergence analysis of stochastic adaptive linear filtering
• Stabilization of Linear Systems by Noise
• Tracking properties of adaptive signal processing algorithms

Cited in

(5)

• How exciting can a signal really be?
• Local minima escape transients by stochastic gradient descent algorithms in blind adaptive equalizers
• Exponential convergence of the Kalman filter based parameter estimation algorithm
• scientific article; zbMATH DE number 4120105 (Why is no real title available?)
• Adaptation and tracking in system identification - a survey

This page was built for publication: Convergence rate determination for gradient-based adaptive estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1084367)