Bayesian Estimation for Nonstandard Loss Functions Using a Parametric Family of Estimators

Bayesian estimation with other loss functions than the standard hit-or-miss loss or the quadratic loss often yields optimal Bayesian estimators (OBEs) that can only be formulated as optimization problems and which have to be solved for each new observation. The contribution of this paper is to introduce a new parametric family of estimators to circumvent this problem. By restricting the estimator to lie in this family, we split the estimation problem into two parts: In a first step, we have to find the best estimator with respect to the Bayes risk for a given nonstandard loss function, which has to be done only once. The second step then calculates the estimate for an observation using importance sampling. The computational complexity of this second step is therefore comparable to that of an MMSE estimator if the MMSE estimator also uses Monte Carlo integration. We study the proposed parametric family using two examples and show that the estimator family gives for both a good approximation of the OBE.

[1]  Bruno O. Shubert,et al.  Random variables and stochastic processes , 1979 .

[2]  Jan Gerhard Norstrøm,et al.  The use of precautionary loss functions in risk analysis , 1996, IEEE Trans. Reliab..

[3]  David Malah,et al.  Speech enhancement using a minimum mean-square error log-spectral amplitude estimator , 1984, IEEE Trans. Acoust. Speech Signal Process..

[4]  David A. Binder,et al.  Approximations to Bayesian clustering rules , 1981 .

[5]  Martin S. Levy,et al.  BLINEX: A BOUNDED ASYMMETRIC LOSS FUNCTION WITH APPLICATION TO BAYESIAN ESTIMATION , 2001 .

[6]  Ephraim.  Speech enhancement using a minimum mean square error short-time spectral amplitude estimator , 1984 .

[7]  C. Robert,et al.  Estimating Mixtures of Regressions , 2003 .

[8]  Alan C. Bovik,et al.  Mean squared error: Love it or leave it? A new look at Signal Fidelity Measures , 2009, IEEE Signal Processing Magazine.

[9]  Lawrence D. Brown,et al.  INADMISSIBILITY OF THE USUAL ESTIMATORS OF SCALE PARAMETERS IN PROBLEMS WITH UNKNOWN LOCATION AND SCALE PARAMETERS , 1968 .

[10]  Philipos C. Loizou,et al.  Speech enhancement based on perceptually motivated bayesian estimators of the magnitude spectrum , 2005, IEEE Transactions on Speech and Audio Processing.

[11]  Steven Kay,et al.  Fundamentals Of Statistical Signal Processing , 2001 .

[12]  Milton Abramowitz,et al.  Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables , 1964 .

[13]  A. Zellner.  Bayesian Estimation and Prediction Using Asymmetric Loss Functions , 1986 .

[14]  S. Kay.  Fundamentals of statistical signal processing: estimation theory , 1993 .

[15]  Jun S. Liu,et al.  Monte Carlo strategies in scientific computing , 2001 .

[16]  Christian P. Robert,et al.  The Bayesian choice : from decision-theoretic foundations to computational implementation , 2007 .

[17]  John G. Proakis,et al.  Probability, random variables and stochastic processes , 1985, IEEE Trans. Acoust. Speech Signal Process..

[18]  Martin Pincus,et al.  Letter to the Editor - -A Closed Form Solution of Certain Programming Problems , 1968, Oper. Res..

[19]  M. Abramowitz,et al.  Handbook of Mathematical Functions with Formulas, Graphs and Mathematical Tables. U.S. Department of Commerce, National Bureau of Standards , 1965 .

[20]  Robert W. Heath,et al.  Design of Linear Equalizers Optimized for the Structural Similarity Index , 2008, IEEE Transactions on Image Processing.

[21]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[22]  D. Owen.  Handbook of Mathematical Functions with Formulas , 1965 .

[23]  Eric Plourde,et al.  Auditory-Based Spectral Amplitude Estimators for Speech Enhancement , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[24]  D. Binder.  Bayesian cluster analysis , 1978 .

[25]  Eric Plourde,et al.  Generalized Bayesian Estimators of the Spectral Amplitude for Speech Enhancement , 2009, IEEE Signal Processing Letters.

[26]  John K Kruschke,et al.  Bayesian data analysis. , 2010, Wiley interdisciplinary reviews. Cognitive science.

[27]  Z. Porosiński,et al.  On robust Bayesian estimation under some asymmetric and bounded loss function , 2009 .

[28]  C. Robert,et al.  Computational and Inferential Difficulties with Mixture Posterior Distributions , 2000 .

[29]  Yi Hu,et al.  Evaluation of Objective Quality Measures for Speech Enhancement , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[30]  Methods for objective and subjective assessment of quality Perceptual evaluation of speech quality ( PESQ ) : An objective method for end-to-end speech quality assessment of narrow-band telephone networks and speech codecs , 2002 .

[31]  Håvard Rue,et al.  New Loss Functions in Bayesian Imaging , 1995 .

[32]  C. Robert.  Simulation of truncated normal variables , 2009, 0907.4010.

[33]  P. Green,et al.  Bayesian Model-Based Clustering Procedures , 2007 .

[34]  Bin Yang,et al.  A parametric family of Bayesian estimators for non-standard loss functions , 2010, 2010 18th European Signal Processing Conference.

[35]  C. Stein.  Inadmissibility of the usual estimator for the variance of a normal distribution with unknown mean , 1964 .

[36]  Philipos C. Loizou.  Speech Enhancement Algorithms: A Survey , 2011 .

[37]  Richard M. Schwartz,et al.  Enhancement of speech corrupted by acoustic noise , 1979, ICASSP.

[38]  Susanto Rahardja,et al.  /spl beta/-order MMSE spectral amplitude estimation for speech enhancement , 2005, IEEE Transactions on Speech and Audio Processing.