Tom Burr
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Tom Burr has a total of 149 co-authors. Here are the top co-authors by number of shared publications:
- Michael S. Hamada
Statistical Sciences Group, Los Alamos National Laboratory, Los Alamos, New Mexico
7 shared publications Michael S. Hamada
6 shared publications- Kary Myers
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
5 shared publications M. S. Hamada
Statistical Sciences Group Los Alamos National Laboratory Los Alamos NM 87545 USA
5 shared publications- Nicolas Hengartner
Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, NM 87545
5 shared publications Brian Weaver
3 shared publications
Latest Publications
- Open Access
- 25 December 2025
- Published by MDPI AG in Algorithms
- Vol. 19 (1) , 25
Variance Preserving Spectral Subsampling
Generating statistically faithful short-duration gamma-ray spectra from a single long measurement is essential in nuclear safeguards, supporting tasks such as algorithm development and machine-learning applications, especially when list-mode data are unavailable. Existing subsampling methods often distort the statistical characteristics of genuine short-duration measurements, leading to biased or unreliable analytical outcomes and thereby undermining downstream tasks. In this work, we compare five subsampling approaches using a benchmark set of 156 genuine replicate spectra collected with a high-purity germanium detector. We evaluate each method with respect to run-to-run variance, channel-to-channel variance, and preservation of total counts (losslessness). Across a wide range of subsampling ratios, only binomial subsampling without replacement consistently reproduces the statistical properties of genuine short-duration spectra, maintaining proper dispersion even in sparse spectral regions and perfectly preserving total counts. These results provide a mathematically principled and practically validated framework for generating synthetically shortened spectra when true short-duration measurements are unavailable.
- 25 December 2025
- Published by MDPI AG in Algorithms
- Vol. 19 (1) , 25
Open AccessVariance Preserving Spectral Subsampling
Generating statistically faithful short-duration gamma-ray spectra from a single long measurement is essential in nuclear safeguards, supporting tasks such as algorithm development and machine-learning applications, especially when list-mode data are unavailable. Existing subsampling methods often distort the statistical characteristics of genuine short-duration measurements, leading to biased or unreliable analytical outcomes and thereby undermining downstream tasks. In this work, we compare five subsampling approaches using a benchmark set of 156 genuine replicate spectra collected with a high-purity germanium detector. We evaluate each method with respect to run-to-run variance, channel-to-channel variance, and preservation of total counts (losslessness). Across a wide range of subsampling ratios, only binomial subsampling without replacement consistently reproduces the statistical properties of genuine short-duration spectra, maintaining proper dispersion even in sparse spectral regions and perfectly preserving total counts. These results provide a mathematically principled and practically validated framework for generating synthetically shortened spectra when true short-duration measurements are unavailable.
- Open Access
- 18 January 2023
- Published by IntechOpen
Change Detection by Monitoring Residuals from Time Series Models
Change detection in time series can be approached by fitting a model to the no-change, ordinary background data and then monitoring time series of residuals, where a residual is defined as residual = data – fit. In many applications, models that fit time series data lead to residuals that exhibit no patterns unless the signal of interest is present. Therefore, an effective signal or change detection approach is to first fit a time series model to the background data without any signal and then monitor the time series of residuals for evidence of the signal. This chapter briefly reviews a few time series modeling options and then focuses on statistical tests for monitoring residuals, including Page’s cumulative sum (cusum, a type of scan statistic), the ordinary cumulative sum (cumsum), the matched filter (a version of the Neyman-Pearson test statistic), and pattern tests, such as those used in quality control. Simulation and analytical approximation methods are recommended for studying test behavior, as illustrated in three examples.
- 18 January 2023
- Published by IntechOpen
Open AccessChange Detection by Monitoring Residuals from Time Series Models
Change detection in time series can be approached by fitting a model to the no-change, ordinary background data and then monitoring time series of residuals, where a residual is defined as residual = data – fit. In many applications, models that fit time series data lead to residuals that exhibit no patterns unless the signal of interest is present. Therefore, an effective signal or change detection approach is to first fit a time series model to the background data without any signal and then monitor the time series of residuals for evidence of the signal. This chapter briefly reviews a few time series modeling options and then focuses on statistical tests for monitoring residuals, including Page’s cumulative sum (cusum, a type of scan statistic), the ordinary cumulative sum (cumsum), the matched filter (a version of the Neyman-Pearson test statistic), and pattern tests, such as those used in quality control. Simulation and analytical approximation methods are recommended for studying test behavior, as illustrated in three examples.
- 19 December 2022
- Published by Elsevier in Nuclear Engineering and Design
- Vol. 402
Impact of measurement error on deep neural networks for nuclear material accountancy
- 19 December 2022
- Published by Elsevier in Nuclear Engineering and Design
- Vol. 402
Impact of measurement error on deep neural networks for nuclear material accountancy
- Open Access
- 24 November 2021
- Published by MDPI AG in Algorithms
- Vol. 14 (12) , 340
Overview of Algorithms for Using Particle Morphology in Pre-Detonation Nuclear Forensics
A major goal in pre-detonation nuclear forensics is to infer the processing conditions and/or facility type that produced radiological material. This review paper focuses on analyses of particle size, shape, texture (“morphology”) signatures that could provide information on the provenance of interdicted materials. For example, uranium ore concentrates (UOC or yellowcake) include ammonium diuranate (ADU), ammonium uranyl carbonate (AUC), sodium diuranate (SDU), magnesium diuranate (MDU), and others, each prepared using different salts to precipitate U from solution. Once precipitated, UOCs are often dried and calcined to remove adsorbed water. The products can be allowed to react further, forming uranium oxides UO3, U3O8, or UO2 powders, whose surface morphology can be indicative of precipitation and/or calcination conditions used in their production. This review paper describes statistical issues and approaches in using quantitative analyses of measurements such as particle size and shape to infer production conditions. Statistical topics include multivariate t tests (Hotelling’s ), design of experiments, and several machine learning (ML) options including decision trees, learning vector quantization neural networks, mixture discriminant analysis, and approximate Bayesian computation (ABC). ABC is emphasized as an attractive option to include the effects of model uncertainty in the selected and fitted forward model used for inferring processing conditions.- 24 November 2021
- Published by MDPI AG in Algorithms
- Vol. 14 (12) , 340
Open AccessOverview of Algorithms for Using Particle Morphology in Pre-Detonation Nuclear Forensics
A major goal in pre-detonation nuclear forensics is to infer the processing conditions and/or facility type that produced radiological material. This review paper focuses on analyses of particle size, shape, texture (“morphology”) signatures that could provide information on the provenance of interdicted materials. For example, uranium ore concentrates (UOC or yellowcake) include ammonium diuranate (ADU), ammonium uranyl carbonate (AUC), sodium diuranate (SDU), magnesium diuranate (MDU), and others, each prepared using different salts to precipitate U from solution. Once precipitated, UOCs are often dried and calcined to remove adsorbed water. The products can be allowed to react further, forming uranium oxides UO3, U3O8, or UO2 powders, whose surface morphology can be indicative of precipitation and/or calcination conditions used in their production. This review paper describes statistical issues and approaches in using quantitative analyses of measurements such as particle size and shape to infer production conditions. Statistical topics include multivariate t tests (Hotelling’s ), design of experiments, and several machine learning (ML) options including decision trees, learning vector quantization neural networks, mixture discriminant analysis, and approximate Bayesian computation (ABC). ABC is emphasized as an attractive option to include the effects of model uncertainty in the selected and fitted forward model used for inferring processing conditions. - 1 April 2021
- Published by Elsevier in Chemometrics and Intelligent Laboratory Systems
- Vol. 211, 104224
Bottom-up and top-down uncertainty quantification for measurements
- 1 April 2021
- Published by Elsevier in Chemometrics and Intelligent Laboratory Systems
- Vol. 211, 104224
Bottom-up and top-down uncertainty quantification for measurements
- Open Access
- 19 October 2020
- Published by MDPI AG in Algorithms
- Vol. 13 (10) , 265
Application of the Approximate Bayesian Computation Algorithm to Gamma-Ray Spectroscopy
Radioisotope identification (RIID) algorithms for gamma-ray spectroscopy aim to infer what isotopes are present and in what amounts in test items. RIID algorithms either use all energy channels in the analysis region or only energy channels in and near identified peaks. Because many RIID algorithms rely on locating peaks and estimating each peak’s net area, peak location and peak area estimation algorithms continue to be developed for gamma-ray spectroscopy. This paper shows that approximate Bayesian computation (ABC) can be effective for peak location and area estimation. Algorithms to locate peaks can be applied to raw or smoothed data, and among several smoothing options, the iterative bias reduction algorithm (IBR) is recommended; the use of IBR with ABC is shown to potentially reduce uncertainty in peak location estimation. Extracted peak locations and areas can then be used as summary statistics in a new ABC-based RIID. ABC allows for easy experimentation with candidate summary statistics such as goodness-of-fit scores and peak areas that are extracted from relatively high dimensional gamma spectra with photopeaks (1024 or more energy channels) consisting of count rates versus energy for a large number of gamma energies.
- 19 October 2020
- Published by MDPI AG in Algorithms
- Vol. 13 (10) , 265
Open AccessApplication of the Approximate Bayesian Computation Algorithm to Gamma-Ray Spectroscopy
Radioisotope identification (RIID) algorithms for gamma-ray spectroscopy aim to infer what isotopes are present and in what amounts in test items. RIID algorithms either use all energy channels in the analysis region or only energy channels in and near identified peaks. Because many RIID algorithms rely on locating peaks and estimating each peak’s net area, peak location and peak area estimation algorithms continue to be developed for gamma-ray spectroscopy. This paper shows that approximate Bayesian computation (ABC) can be effective for peak location and area estimation. Algorithms to locate peaks can be applied to raw or smoothed data, and among several smoothing options, the iterative bias reduction algorithm (IBR) is recommended; the use of IBR with ABC is shown to potentially reduce uncertainty in peak location estimation. Extracted peak locations and areas can then be used as summary statistics in a new ABC-based RIID. ABC allows for easy experimentation with candidate summary statistics such as goodness-of-fit scores and peak areas that are extracted from relatively high dimensional gamma spectra with photopeaks (1024 or more energy channels) consisting of count rates versus energy for a large number of gamma energies.
- 20 June 2019
- Published by Office of Scientific and Technical Information (OSTI)
Measurement Error Variance Estimation in Gamma-Spectroscopy Data using FRAM
The U.S. Department of Energy's Office of Scientific and Technical Information
- 20 June 2019
- Published by Office of Scientific and Technical Information (OSTI)
Measurement Error Variance Estimation in Gamma-Spectroscopy Data using FRAM
The U.S. Department of Energy's Office of Scientific and Technical Information
- Open Access
Setting Alarm Thresholds in Measurements with Systematic and Random Errors
For statistical evaluations that involve within-group and between-group variance components (denoted and , respectively), there is sometimes a need to monitor for a shift in the mean of time-ordered data. Uncertainty in the estimates and should be accounted for when setting alarm thresholds to check for a mean shift as both and must be estimated. One-way random effects analysis of variance (ANOVA) is the main tool for analysing such grouped data. Nearly all of the ANOVA applications assume that both the within-group and between-group components are normally distributed. However, depending on the application, the within-group and/or between-group probability distributions might not be well approximated by a normal distribution. This review paper uses the same example throughout to illustrate the possible approaches to setting alarm limits in grouped data, depending on what is assumed about the within-group and between-group probability distributions. The example involves measurement data, for which systematic errors are assumed to remain constant within a group, and to change between groups. The false alarm probability depends on the assumed measurement error model and its within-group and between-group error variances, which are estimated while using historical data, usually with ample within-group data, but with a small number of groups (three to 10 typically). This paper illustrates the parametric, semi-parametric, and non-parametric options to setting alarm thresholds in such grouped data.Open AccessSetting Alarm Thresholds in Measurements with Systematic and Random Errors
For statistical evaluations that involve within-group and between-group variance components (denoted and , respectively), there is sometimes a need to monitor for a shift in the mean of time-ordered data. Uncertainty in the estimates and should be accounted for when setting alarm thresholds to check for a mean shift as both and must be estimated. One-way random effects analysis of variance (ANOVA) is the main tool for analysing such grouped data. Nearly all of the ANOVA applications assume that both the within-group and between-group components are normally distributed. However, depending on the application, the within-group and/or between-group probability distributions might not be well approximated by a normal distribution. This review paper uses the same example throughout to illustrate the possible approaches to setting alarm limits in grouped data, depending on what is assumed about the within-group and between-group probability distributions. The example involves measurement data, for which systematic errors are assumed to remain constant within a group, and to change between groups. The false alarm probability depends on the assumed measurement error model and its within-group and between-group error variances, which are estimated while using historical data, usually with ample within-group data, but with a small number of groups (three to 10 typically). This paper illustrates the parametric, semi-parametric, and non-parametric options to setting alarm thresholds in such grouped data. - Open Access
- 1 January 2019
- Published by Wiley in Science and Technology of Nuclear Installations
- Vol. 2019, 1-10
Analysis of Variance for Item Differences in Verification Data with Unknown Groups
For sequentially collected data, this paper introduces a lag-one differencing method to estimate the random error standard deviation and then uses the estimate to calculate a change detection threshold in a moving window method to detect shifts in the short-term systematic error. Performance results on simulated and real data are presented. Fortunately, the impact of having to perform change detection on the estimated short-term systematic and random error variances is anticipated to be modest or small. The motivating example arises from facilities under nuclear safeguards agreements, where inspector data collected during International Atomic Energy Agency (IAEA) verifications are compared to corresponding operator data. The differences between the operator and inspector values are evaluated using an application of analysis of variance (ANOVA). Typically, it is assumed that short-term systematic errors change across inspection periods, so inspection periods form the groups used in the ANOVA. In some data sets, it appears that the short-term errors have changed at other times, so change detection methods could be used to detect the actual change times.- 1 January 2019
- Published by Wiley in Science and Technology of Nuclear Installations
- Vol. 2019, 1-10
Open AccessAnalysis of Variance for Item Differences in Verification Data with Unknown Groups
For sequentially collected data, this paper introduces a lag-one differencing method to estimate the random error standard deviation and then uses the estimate to calculate a change detection threshold in a moving window method to detect shifts in the short-term systematic error. Performance results on simulated and real data are presented. Fortunately, the impact of having to perform change detection on the estimated short-term systematic and random error variances is anticipated to be modest or small. The motivating example arises from facilities under nuclear safeguards agreements, where inspector data collected during International Atomic Energy Agency (IAEA) verifications are compared to corresponding operator data. The differences between the operator and inspector values are evaluated using an application of analysis of variance (ANOVA). Typically, it is assumed that short-term systematic errors change across inspection periods, so inspection periods form the groups used in the ANOVA. In some data sets, it appears that the short-term errors have changed at other times, so change detection methods could be used to detect the actual change times. - Open Access
- 21 December 2018
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1141 (1) , 012127
Approximate Bayesian Computation applied to Metrology for Nuclear Safeguards Journal of Physics: Conference Series
Approximate Bayesian Computation (ABC) is an inference option if a likelihood for the measurement data is not available, but a forward model is available that outputs predicted observables for specified input parameters. This paper applies ABC to the enrichment meter principle (EMP) method, which is used in nuclear safeguards and does not have an explicit likelihood. A key aspect of metrology is uncertainty quantification (UQ), approached from physical first principles ("bottom-up") or approached empirically by comparing measurements from different methods and/or laboratories ("top-down"). Although ABC is not commonly used in metrology, the EMP example illustrates advantages in ABC compared to current bottom-up approaches. ABC is also shown to be useful in top-down UQ. As a diagnostic, in bottom-up and top-down applications of ABC, the actual coverages of probability intervals are compared to the true coverages. If an ABC-based interval for a parameter is constructed to contain approximately 95% of the true parameter values, then it is important to verify that the actual coverage is close to 95%. It is shown that one advantage of ABC compared to other Bayesian approaches is its apparent robustness to miss-specifying the likelihood while maintaining good agreement between nominal and actual coverage.
- 21 December 2018
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1141 (1) , 012127
Open AccessApproximate Bayesian Computation applied to Metrology for Nuclear Safeguards Journal of Physics: Conference Series
Approximate Bayesian Computation (ABC) is an inference option if a likelihood for the measurement data is not available, but a forward model is available that outputs predicted observables for specified input parameters. This paper applies ABC to the enrichment meter principle (EMP) method, which is used in nuclear safeguards and does not have an explicit likelihood. A key aspect of metrology is uncertainty quantification (UQ), approached from physical first principles ("bottom-up") or approached empirically by comparing measurements from different methods and/or laboratories ("top-down"). Although ABC is not commonly used in metrology, the EMP example illustrates advantages in ABC compared to current bottom-up approaches. ABC is also shown to be useful in top-down UQ. As a diagnostic, in bottom-up and top-down applications of ABC, the actual coverages of probability intervals are compared to the true coverages. If an ABC-based interval for a parameter is constructed to contain approximately 95% of the true parameter values, then it is important to verify that the actual coverage is close to 95%. It is shown that one advantage of ABC compared to other Bayesian approaches is its apparent robustness to miss-specifying the likelihood while maintaining good agreement between nominal and actual coverage.