Bayesian spectral deconvolution with the exchange Monte Carlo method

An analytical method to deconvolute spectral data into a number of simple bands is extremely important in the analysis of the chemical properties of matter. However, there are two fundamental problems with such deconvolution methods. One is how to determine the number of bands without resorting to heuristics. The other is difficulty in avoiding the parameter solution trapped into local minima due to the hierarchy and the nonlinearity of the system. In this study, we propose a novel method of spectral deconvolution based on Bayesian estimation with the exchange Monte Carlo method, which is an application of the integral approximation of stochastic complexity and the exchange Monte Carlo method. We also experimentally show its effectiveness on synthetic data and on reflectance spectral data of olivine, one of the most common minerals of terrestrial planets.

[1]  Narasimhan Sundararajan,et al.  A generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation , 2005, IEEE Transactions on Neural Networks.

[2]  I ScottKirkpatrick.  Optimization by Simulated Annealing: Quantitative Studies , 1984 .

[3]  Sumio Watanabe,et al.  Equations of States in Singular Statistical Estimation , 2007, Neural Networks.

[4]  Mark Jerrum,et al.  Polynomial-Time Approximation Algorithms for the Ising Model , 1990, SIAM J. Comput..

[5]  P. S. Lewis,et al.  Function approximation and time series prediction with neural networks , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[6]  Ajay Jasra,et al.  Population-Based Reversible Jump Markov Chain Monte Carlo , 2007, 0711.0186.

[7]  K. Hukushima,et al.  Exchange Monte Carlo Method and Application to Spin Glass Simulations , 1995, cond-mat/9512035.

[8]  Hirotugu Akaike,et al.  Likelihood and the Bayes procedure , 1980 .

[9]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[10]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .

[11]  J. Griffin,et al.  Time-domain Bayesian detection and estimation of noisy damped sinusoidal signals applied to NMR spectroscopy. , 2007, Journal of magnetic resonance.

[12]  C. I. Mosier.  I. Problems and Designs of Cross-Validation 1 , 1951 .

[13]  Sumio Watanabe,et al.  Asymptotic behavior of exchange ratio in exchange Monte Carlo method , 2008, Neural Networks.

[14]  M. J. D. Powell,et al.  Radial basis functions for multivariable interpolation: a review , 1987 .

[15]  Sumio Watanabe,et al.  Exchange Monte Carlo Sampling From Bayesian Posterior for Singular Learning Machines , 2008, IEEE Transactions on Neural Networks.

[16]  Carle M. Pieters,et al.  Deconvolution of mineral absorption bands: An improved approach , 1990 .

[17]  Y. Ogata.  A Monte Carlo method for an objective Bayesian procedure , 1990 .

[18]  Scott Kirkpatrick,et al.  Optimization by simulated annealing: Quantitative studies , 1984 .

[19]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[20]  T. J. Kennett,et al.  Bayesian deconvolution I: Convergent properties , 1978 .

[21]  John C. Platt.  A Resource-Allocating Network for Function Interpolation , 1991, Neural Computation.

[22]  Carle M. Pieters,et al.  Determining the composition of olivine from reflectance spectroscopy , 1998 .

[23]  C. Andrieu,et al.  Bayesian model selection and parameter estimation of nuclear emission spectra using RJMCMC , 2003 .