Title :
A Dynamic Parzen Window Approach Based on Error-entropy Minimization Algorithm for Supervised Training of Nonlinear Adaptive System
Author :
Zibin, Wang ; Xuemei Ren ; Yan, Xuemei Liu
Author_Institution :
Beijing Inst. of Technol., Beijing
Abstract :
This paper presents a dynamic Parzen window estimator in the MEE approach for supervised training of nonlinear adaptive system. By adjusting the Parzen window width dynamically so that the overall information force (OIF) among error-samples of each step is as large as possible, the training speed is accelerated and the error is reduced. The simulation result has proved the effectiveness and robustness of this algorithm.
Keywords :
adaptive systems; entropy; learning (artificial intelligence); minimisation; nonlinear dynamical systems; dynamic Parzen window estimator; error-entropy minimization algorithm; nonlinear adaptive system; overall information force; Adaptive systems; Control systems; Data mining; Entropy; Error correction; Kernel; Mean square error methods; Minimization methods; Nonlinear dynamical systems; Probability density function; Dynamic Parzen window approach; Error-entropy minimization (MEE); Information Theoretic Learning (ITL);
Conference_Titel :
Control Conference, 2007. CCC 2007. Chinese
Conference_Location :
Hunan
Print_ISBN :
978-7-81124-055-9
Electronic_ISBN :
978-7-900719-22-5
DOI :
10.1109/CHICC.2006.4347162
