1. Alexander M. Venzin
PRESENT ADDRESS
8 Mayerhofgasse Apt 16
Vienna, Austria 1040
+43 (676) 402-0195
PERMANENT ADDRESS
414 Goebel Dr
Frankfort, KY 40601
(502) 695-3464
EXPERIENCE
Junior Professional Officer / P-2
International Atomic Energy Agency, Vienna, Austria
June 2015 - Present
• Organizing, designing, and analyzing interlaboratory comparison exercises in the domains of
bulk analysis, particle analysis, and elemental impurities characterization. These exercises
are organized and analyzed with respect to the international standards (ISO, EURACHEM,
and IUPAC).
• Developing and publishing a library of statistical applications for use in analytical chemistry.
The library covers topics in ISO 13528 (Guide to conducting proficiency test exercises), un-
certainty propagation, robust parameter estimation, and routines and visualization tools for
random effects ANOVA models. This is a fully documented R programming package with
additional C++ components.
• Comprehension and application of ISO guides 35 (production of reference materials), 98 (un-
certainty modeling), 5725 (statistical models in metrology), 7870 (quality control charting),
13528 (guide to proficiency testing), 21748 (more uncertainty modeling), and the JCGM 100
(the GUM) to routine laboratory procedures.
• Building data acquisition and classification algorithms for scanning electron microscope appli-
cations. In particular, these applications are interested in using x-ray spectra measurements
to build an unsupervised classifier.
• Direct communication with management from the IAEA’s network of analytical laboratories
(NWAL). This network spans at least 10 nationalities and requires a broad understanding of
cultural differences and the ability to communicate with a wide range of different language
skills.
Research Associate
Pacific Northwest National Laboratory, Applied Statistics and Computational Modeling Group,
Richland, WA
March 2013 - May 2015
• Developed numerical algorithms for the stratified compliance sampling (SCS) and combined
judgmental and random (CJR) sampling models for PNNL’s Visual Sample Plan software.
These are Bayesian hierarchical models built on the Beta-binomial conjugate distribution.
http://vsp.pnnl.gov/
• Worked with a team modeling power markets for PJM Interconnection using the Bayes’
Model Averaging (BMA) algorithm. PJM interconnection is the largest regional transmission
organization in the United States. This software significantly outperformed the pre-existing
neural network in use by PJM interconnection. The project won an R&D 100 Award.
• Internally crowd funded to develop R software for simple large scale parallel computation.
This package automated the process of cluster queuing and allocation for users and reduced
2. the complexity to a modified version of R’s “apply” (for loop) syntax. The software also
provided automated Hadoop file server creation.
• Built a feature extraction algorithm from a Diabetes Antibody Standardization Program
(DASP) data set. The data set was composed of a large collection of proteomics, metabolomics,
and lipidomics measurements taken from a subset of individuals that either had already de-
veloped or were at risk of developing type 1 Diabetes. The features of interest were the subset
of omics data that could correctly predict whether an individual had type 1 Diabetes.
• Built a feature extractor and classifier for omics data collected from two populations of
Yersinia Pestis (black plague). The study was exploratory in nature and the goal was to
find a set of predictors that allows you to uniquely identify whether or not the YP in question
originated in the wild or was grown in a laboratory.
Graduate Research Assistant
Air Force Institute of Technology, Applied Mathematics Department, Wright-Patterson Air Force
Base, OH
March 2011 - March 2013
• Studied under the supervision of Christine Schubert-Kabban PhD and Mark Oxley PhD. My
thesis revolved around quantifying the bias in “fused” (probability transformation) Receiver
Operating Characteristic (ROC) curves. ROC curves are used in statistical classification to
quantify accuracy.
• Presented work for the Department of Homeland Security’s Explosives Division using my
thesis work to demonstrate how sensor information from multiple non-destructive instruments
can be used to improve the trade-off between true positives and false positives when classifying
hidden explosives material. This was part of Pacific Northwest National Laboratory’s summer
graduate program.
TECHNICAL SKILLS
Statistical Models: Mixed Effects Models, ANOVA, Propagation of Uncertainty (GUM), GLS, Pe-
nalized regression, Bayesian Conjugate models, Regression Trees / Forests, SVM, logistic regression,
ANN, Spatial Poisson Processes, numerical quadrature methods, Monte Carlo Methods, Sampling
methodology.
Computer Languages: R, C/C++, Python, LATEX, Bash
Databases: SQL/SQLite
General Skills: MS Office tools (Word, Excel, Powerpoint, Access, Outlook), Linux, Unix
AWARDS
• 2015 R&D 100 Award - Power Model Integrator: A System for More Accurate Energy Forecasts.
http://www.pnl.gov/about/rd100awards.asp
• Outstanding Performance Award - May 2015 (PNNL)
PEER REVIEWED PUBLICATIONS
• Robust estimation of variance components as described in international standards and comparison
to a robust linear mixed models approach. Venzin, Alexander M., and Walsh, Stephen J. In
preparation, targeting Technometrics.
3. • Signatures for Mass Spectrometry Data Quality. Amidan, Brett G., Orton, Daniel J., LaMarche,
Brian L., Monroe, Matthew E., Moore, Ronald J., Venzin, Alexander M., Smith, Richard D.,
Sego, Landon H., Tardiff, Mark F., Payne, Samuel H. Journal of Proteome Research. Vol 13. Issue
4. Pages 2215-2222.
CONFERENCE PRESENTATIONS
• Using reproducibility to test the adequacy of GUM based Uncertainty Quantification (Peer re-
viewed). Walsh, Stephen J., Venzin, Alexander M., Wegrzynek, Dariusz, Mansoux, Catherine.
American Nuclear Society - Advances in Nuclear Nonproliferation Technology and Policy Confer-
ence 2016.
• An Investigation into Label Fusion on Sparse Data. Venzin, Alexander M., Oxley, Mark E.,
Beagley, Nathaniel. Signature Discovery Workshop IEEE ISI Conference 2013.
TECHNICAL REPORTS
• 2015 US DOE Sponsored Proficiency Test for Bulk Analysis of Environmental Swipe Samples.
Venzin, Alexander M., Walsh, Stephen J. SG-RP-14132.
• 2016 blind inter-laboratory comparison for determination of low level Pu in environmental samples
by bulk analysis. Venzin, Alexander M., Walsh, Stephen J. SG-RP-14060.
• IAEA 2015 inter-laboratory comparison - determination of elemental impurities in uranium ox-
ide matrix materials. Walsh, Stephen J., Venzin, Alexander M., Bulhya, Sergei, Macsik Z.,
Wegrzynek, Dariusz, Penkin, Maxim, Mansoux, Catherine, Martin, Paul, Kuno, Yusuke. SG-RP-
13965.
• Consensus value Estimation for ES Blank Swipes from Lot #49F3C. Venzin, Alexander M.,
Walsh, Stephen J. SG-RP-13670.
• 2015 Nuclear Material Proficiency Test. Walsh, Stephen J., Wegrzynek, Dariusz, Venzin, Alexan-
der M., Repinc, Urska, Schoenfeld, Julia, Amaraggi, David, Balsley, Steve. SG-RP-13613.
EDUCATION
Air Force Institute of Technology, Wright-Patterson Air Force Base, OH 45433
M.S. Applied Mathematics, September 2012
GPA 3.67/4.00
Western Kentucky University, Bowling Green, KY 42101
B.S. Mathematics & Chemistry, May 2010
GPA 3.88/4.00
ADDITIONAL INFORMATION
Source code and publications are available upon request.