skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Preliminary Results on Uncertainty Quantification for Pattern Analytics

Abstract

This report summarizes preliminary research into uncertainty quantification for pattern ana- lytics within the context of the Pattern Analytics to Support High-Performance Exploitation and Reasoning (PANTHER) project. The primary focus of PANTHER was to make large quantities of remote sensing data searchable by analysts. The work described in this re- port adds nuance to both the initial data preparation steps and the search process. Search queries are transformed from does the specified pattern exist in the data? to how certain is the system that the returned results match the query? We show example results for both data processing and search, and discuss a number of possible improvements for each.

Authors:
 [1];  [1];  [1];  [1];  [1];  [1];  [1];  [1]
  1. Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
Publication Date:
Research Org.:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Sandia National Lab. (SNL-CA), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1221712
Report Number(s):
SAND2015-8099
603967
DOE Contract Number:
AC04-94AL85000
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English

Citation Formats

Stracuzzi, David John, Brost, Randolph, Chen, Maximillian Gene, Malinas, Rebecca, Peterson, Matthew Gregor, Phillips, Cynthia A., Robinson, David G., and Woodbridge, Diane. Preliminary Results on Uncertainty Quantification for Pattern Analytics. United States: N. p., 2015. Web. doi:10.2172/1221712.
Stracuzzi, David John, Brost, Randolph, Chen, Maximillian Gene, Malinas, Rebecca, Peterson, Matthew Gregor, Phillips, Cynthia A., Robinson, David G., & Woodbridge, Diane. Preliminary Results on Uncertainty Quantification for Pattern Analytics. United States. doi:10.2172/1221712.
Stracuzzi, David John, Brost, Randolph, Chen, Maximillian Gene, Malinas, Rebecca, Peterson, Matthew Gregor, Phillips, Cynthia A., Robinson, David G., and Woodbridge, Diane. Tue . "Preliminary Results on Uncertainty Quantification for Pattern Analytics". United States. doi:10.2172/1221712. https://www.osti.gov/servlets/purl/1221712.
@article{osti_1221712,
title = {Preliminary Results on Uncertainty Quantification for Pattern Analytics},
author = {Stracuzzi, David John and Brost, Randolph and Chen, Maximillian Gene and Malinas, Rebecca and Peterson, Matthew Gregor and Phillips, Cynthia A. and Robinson, David G. and Woodbridge, Diane},
abstractNote = {This report summarizes preliminary research into uncertainty quantification for pattern ana- lytics within the context of the Pattern Analytics to Support High-Performance Exploitation and Reasoning (PANTHER) project. The primary focus of PANTHER was to make large quantities of remote sensing data searchable by analysts. The work described in this re- port adds nuance to both the initial data preparation steps and the search process. Search queries are transformed from does the specified pattern exist in the data? to how certain is the system that the returned results match the query? We show example results for both data processing and search, and discuss a number of possible improvements for each.},
doi = {10.2172/1221712},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Sep 01 00:00:00 EDT 2015},
month = {Tue Sep 01 00:00:00 EDT 2015}
}

Technical Report:

Save / Share:
  • This report outlines techniques for extending benchmark generation products so they support uncertainty quantification by benchmarked systems. We describe how uncertainty quantification requirements can be presented to candidate analytical tools supporting SPARQL. We describe benchmark data sets for evaluating uncertainty quantification, as well as an approach for using our benchmark generator to produce data sets for generating benchmark data sets.
  • This report contains 14 appendices which support the mathematical model developed in Volume 1. (TEM)
  • The present study was intended to evaluate a range of statistical performance measures and to develop and test several new methods applied intensively to one model; however, the objectives and scope were still limited. Determination of the uncertainty in model-derived source/receptor relationships did not appear to be feasible from a comparison of the model predictions with the limited number of observations of deposition currently available. At best, it may be possible to estimate or place bounds on the uncertainty in predictions of deposition and to understand the character of those uncertainties. The performance of a model in predicting deposition andmore » air-quality changes in response to large emission changes, outside the bounds of experience with historical evaluated data, cannot be confidently determined by the present methods. New approaches appear to be needed to provide a better understanding of the character of model errors and to estimate the uncertainties in future predictions within a limited domain, based on a comparison of predictions with historical observations. We hope that the results reported on here will be useful and can be applied to other models, as well as refined by further application with more extensive data. 82 refs., 37 figs., 26 tabs.« less
  • This report documents the results of an FY09 ASC V&V Methods level 2 milestone demonstrating new algorithmic capabilities for mixed aleatory-epistemic uncertainty quantification. Through the combination of stochastic expansions for computing aleatory statistics and interval optimization for computing epistemic bounds, mixed uncertainty analysis studies are shown to be more accurate and efficient than previously achievable. Part I of the report describes the algorithms and presents benchmark performance results. Part II applies these new algorithms to UQ analysis of radiation effects in electronic devices and circuits for the QASPR program.
  • This volume is the fourth in a sequence of reports that document the December 1991 preliminary comparison with 40 CFR 191, Subpart B (the Standard; US EPA, 1985) for the Waste Isolation Pilot Plant (WIPP). The three previous volumes describe the background of the project, the performance-assessment methodology, and the 1991 results (Volume 1); the probability and consequence models used in the calculations (Volume 2); and the reference data base (Volume 3). This volume contains the results of uncertainty and sensitivity analyses conducted using the methodology, modeling system, and data described in the earlier volumes. These analysis provide quantitative andmore » qualitative insights on the relationships between uncertainty in the models and data used in the WIPP performance assessment and the resultant uncertainty in the results of the performance assessment. Uncertainty and sensitivity analysis is an important part of the WIPP performance assessment and contributes to the overall analysis in the following areas: (1) assessment of uncertainty in performance-assessment results that must be used in comparison regulatory standards, (2) identification of modeling areas where reductions in uncertainty can significantly improve the confidence that can be a placed in performance-assessment results, and (3) verification that the models used within the performance-assessment process are operating properly.« less