Searcharxiv⌕ Search

arXiv subjects

Nikolai Gagunashvili

Publications and source records attributed to Nikolai Gagunashvili.

9 recordsLinked to original sources
arXiv2017

Tests for Comparing Weighted Histograms. Review and Improvements

Histograms with weighted entries are used to estimate probability density functions. Computer simulation is the main application of this type of histograms. A review on chi-square tests for comparing weighted histograms is presented in this paper. Improvements to these tests that have a size closer to its nominal value are proposed. Numerical examples are presented for evaluation and demonstration of various applications of the tests.

physics.data-an↗
arXiv2016

Unfolding problem clarification and solution validation

The unfolding problem formulation for correcting experimental data distortions due to finite resolution and limited detector acceptance is discussed. A novel validation of the problem solution is proposed. Attention is drawn to fact that different unfolded distributions may satisfy the validation criteria, in which case a conservative approach using entropy is suggested. The importance of analysis of residuals is demonstrated.

physics.data-an↗
arXiv2015

Chi-square goodness of fit tests for weighted histograms. Review and improvements

Weighted histograms are used for the estimation of probability density functions. Computer simulation is the main domain of application of this type of histogram. A review of chi-square goodness of fit tests for weighted histograms is presented in this paper. Improvements are proposed to these tests that have size more close to its nominal value. Numerical examples are presented in this paper for evaluation of tests and to demonstrate various applications of tests.

physics.data-an↗
arXiv2015

Densities mixture unfolding for data obtained from detectors with finite resolution and limited acceptance

A procedure based on a Mixture Density Model for correcting experimental data for distortions due to finite resolution and limited detector acceptance is presented. Addressing the case that the solution is known to be non-negative, in the approach presented here, the true distribution is estimated by a weighted sum of probability density functions with positive weights and with the width of the densities acting as a regularisation parameter responsible for the smoothness of the result. To obtain better smoothing in less populated regions, the width parameter is chosen inversely proportional to the square root of the estimated density. Furthermore, the non-negative garrotte method is used to find the most economic representation of the solution. Cross-validation is employed to determine the optimal values of the resolution and garrotte parameters. The proposed approach is directly applicable to multidimensional problems. Numerical examples in one and two dimensions are presented to illustrate the procedure.

physics.data-an↗
arXiv2011

Machine learning approach to inverse problem and unfolding procedure

A procedure for unfolding the true distribution from experimental data is presented. Machine learning methods are applied for simultaneous identification of an apparatus function and solving of an inverse problem. A priori information about the true distribution from theory or previous experiments is used for Monte-Carlo simulation of the training sample. The training sample can be used to calculate a transformation from the true distribution to the measured one. This transformation provides a robust solution for an unfolding problem with minimal biases and statistical errors for the set of distributions used to create the training sample. The dimensionality of the solved problem can be arbitrary. A numerical example is presented to illustrate and validate the procedure.

physics.data-an↗
arXiv2010

Classifying extremely imbalanced data sets

Imbalanced data sets containing much more background than signal instances are very common in particle physics, and will also be characteristic for the upcoming analyses of LHC data. Following up the work presented at ACAT 2008, we use the multivariate technique presented there (a rule growing algorithm with the meta-methods bagging and instance weighting) on much more imbalanced data sets, especially a selection of D0 decays without the use of particle identification. It turns out that the quality of the result strongly depends on the number of background instances used for training. We discuss methods to exploit this in order to improve the results significantly, and how to handle and reduce the size of large training sets without loss of result quality in general. We will also comment on how to take into account statistical fluctuation in receiver operation characteristic curves (ROC) for comparing classifier methods.

physics.data-an↗
arXiv2009

Application of the rule-growing algorithm RIPPER to particle physics analysis

A large hadron machine like the LHC with its high track multiplicities always asks for powerful tools that drastically reduce the large background while selecting signal events efficiently. Actually such tools are widely needed and used in all parts of particle physics. Regarding the huge amount of data that will be produced at the LHC, the process of training as well as the process of applying these tools to data, must be time efficient. Such tools can be multivariate analysis -- also called data mining -- tools. In this contribution we present the results for the application of the multivariate analysis, rule growing algorithm RIPPER on a problem of particle selection. It turns out that the meta-methods bagging and cost-sensitivity are essential for the quality of the outcome. The results are compared to other multivariate analysis techniques.

physics.data-an↗