A comparison of artificial neural network performance: The case of neutron/gamma pulse shape discrimination

Pulse shape discrimination is investigated using artificial neural networks, namely linear vector quantization and self organizing maps which are employed for classifying neutron and gamma rays at a variety of energies and for different relative sizes of the training and test sets. While classification performance confirms that both approaches are capable of excellent discrimination, some differences between the approaches are observed: linear vector quantization is particularly accurate in classifying the training set; the self organizing map, on the other hand, demonstrates higher prediction accuracy, with its clustering capabilities rendering it less sensitive to classification errors. Comparisons with existing analytical as well as artificial neural network approaches are made.

[1]  Manuel Sanchez-Raya,et al.  Implementation of a neural network for digital pulse shape analysis on a FPGA for on-line identification of heavy ions , 2012 .

[2]  M. Moszynski,et al.  Comparison of n-γ discrimination by zero-crossing and digital charge comparison methods , 1995 .

[3]  C. Morel,et al.  Digital pulse shape discrimination methods for phoswich detectors , 2005, IEEE Nuclear Science Symposium Conference Record, 2005.

[4]  Glenn F. Knoll,et al.  Digital pulse-shape analyzer based on fast sampling of an integrated charge pulse , 1995 .

[5]  Matthias Weiszflog,et al.  An artificial neural network based neutron-gamma discrimination and pile-up rejection framework for the BC-501 liquid scintillation detector , 2009 .

[6]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[7]  F. D. Brooks.  A scintillation counter with neutron and gamma-ray discriminators , 1959 .

[8]  M. Joyce,et al.  A Wavelet Packet Transform Inspired Method of Neutron-Gamma Discrimination , 2010, IEEE Transactions on Nuclear Science.

[9]  David J. C. MacKay,et al.  Bayesian Interpolation , 1992, Neural Computation.

[10]  Frank J. Lynch,et al.  Time Dependence of Scintillations and the Effect on Pulse-Shape Discrimination , 1968 .

[11]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[12]  Yao Xiao,et al.  Research on comparison of credit risk evaluation models based on SOM and LVQ neural network , 2008, 2008 7th World Congress on Intelligent Control and Automation.

[13]  Anthony J. Peyton,et al.  Digital discrimination of neutrons and γ-rays in liquid scintillators using pulse gradient analysis , 2007 .

[14]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[16]  Michael Heil,et al.  Pulse shape analysis of liquid scintillators for neutron studies , 2002 .

[17]  Tsvi Kuflik,et al.  Supervised Learning for Automatic Classification of Documents using Self-Organizing Maps , 2000, DELOS.

[18]  M. Joyce,et al.  An investigation of the digital discrimination of neutrons and γ rays with organic scintillation detectors using an artificial neural network , 2009 .

[19]  IMPLEMENTATION OF DYNAMIC BIAS FOR NEUTRON-PHOTON PULSE SHAPE DISCRIMINATION BY USING NEURAL NETWORK CLASSIFIERS , 1998 .

[20]  M. Joyce,et al.  A Digital Method for the Discrimination of Neutrons and $\gamma$ Rays With Organic Scintillation Detectors Using Frequency Gradient Analysis , 2010, IEEE Transactions on Nuclear Science.

[21]  G. W. Mcbeth,et al.  Pulse shape discrimination in inorganic and organic scintillators. I , 1971 .

[22]  N. Yildiz,et al.  Neural network consistent empirical physical formula construction for neutron–gamma discrimination in gamma ray tracking , 2013 .

[23]  S. Normand,et al.  Discrimination methods between neutron and gamma rays for boron loaded plastic scintillators , 2002 .

[24]  Z. Chalabi,et al.  Classification of the Medical Images by the Kohonen Network SOM and LVQ , 2008 .

[25]  Tsvi Kuflik,et al.  Supporting user-subjective categorization with self-organizing maps and learning vector quantization , 2005, J. Assoc. Inf. Sci. Technol..

[26]  M. Joyce,et al.  A comparison of four different digital algorithms for pulse-shape discrimination in fast scintillators , 2011 .

[27]  Pulse shape analysis using subspace identification methods and particle identification using neural networks in CsI(Tl) scintillators , 2005, IEEE Nuclear Science Symposium Conference Record, 2005.

[28]  H. Spieler,et al.  A simple pulse-shape discrimination circuit , 1974 .

[29]  Fast digitizing techniques applied to scintillation detectors , 2006 .

[30]  Martin Fodslette Møller,et al.  A scaled conjugate gradient algorithm for fast supervised learning , 1993, Neural Networks.

[31]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.