An adaptive kernel width update for correntropy

Correntropy, as an adaptive criterion of Information Theoretic Learning (ITL), has been successfully used in signal processing and machine learning. How to appropriately select the kernel width of correntropy is a crucial problem in correntropy applications. Existing kernel width selection methods are not suitable enough for this problem. In this paper, we develop an adaptive method for kernel width selection in correntropy. Based on the Middleton's non-Gaussian models, this method utilizes the kurtosis as a ratio to adjust the standard deviation of the prediction error to obtain the kernel width online. The superior performance of the new method has been demonstrated by simulation examples in the noisy frequency doubling and echo cancelation problems.

[1]  A. Bowman.  An alternative method of cross-validation for the smoothing of density estimates , 1984 .

[2]  T. Ajdler,et al.  The Plenacoustic Function and Its Sampling , 2006, IEEE Transactions on Signal Processing.

[3]  Jose C. Principe,et al.  Information Theoretic Learning - Renyi's Entropy and Kernel Perspectives , 2010, Information Theoretic Learning.

[4]  Badong Chen,et al.  Kernel adaptive filtering with maximum correntropy criterion , 2011, The 2011 International Joint Conference on Neural Networks.

[5]  José Carlos Príncipe,et al.  Using Correntropy as a cost function in linear adaptive filters , 2009, 2009 International Joint Conference on Neural Networks.

[6]  David Middleton,et al.  Statistical-Physical Models of Electromagnetic Interference , 1977, IEEE Transactions on Electromagnetic Compatibility.

[7]  José Carlos Príncipe,et al.  Generalized correlation function: definition, properties, and application to blind equalization , 2006, IEEE Transactions on Signal Processing.

[8]  D. W. Scott,et al.  Biased and Unbiased Cross-Validation in Density Estimation , 1987 .

[9]  José Carlos Príncipe,et al.  A Pitch Detector Based on a Generalized Correlation Function , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[10]  P. J. Green,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[11]  José Carlos Príncipe,et al.  Information theoretic learning with adaptive kernels , 2011, Signal Process..

[12]  Weifeng Liu,et al.  Correntropy: Properties and Applications in Non-Gaussian Signal Processing , 2007, IEEE Transactions on Signal Processing.

[13]  Jose C. Principe,et al.  A fixed point update for kernel width adaptation in information theoretic criteria , 2010, 2010 IEEE International Workshop on Machine Learning for Signal Processing.

[14]  M. C. Jones,et al.  A Brief Survey of Bandwidth Selection for Density Estimation , 1996 .

[15]  José Carlos Príncipe,et al.  Nonlinear Component Analysis Based on Correntropy , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.