Skip to main content

Paul Diaz

Postdoctoral Researcher-Computational Sciences

| 303-384-6373

Paul Diaz is a GAANN Graduate Fellow. Paul studies uncertainty quantification (UQ) and data-driven modeling. His research focuses on techniques for developing high-accuracy surrogate models from limited noisy data, data analysis and visualization, and explainable uncertainty through global sensitivity analysis. While at NREL, Paul has specialized in applying UQ methods for global sensitivity analysis and constructing surrogate models related to large-scale electrical capacity expansion planning problems. 

Research Interests

Active subspaces for parameter space dimension reduction and data visualization

Sparse polynomial chaos expansions

Compressed sensing and signal processing

Model verification and validation

Ill-posed regression problems

UQ via global sensitivity analysis

Data-driven modeling of dynamical systems

Education

Ph.D., Aerospace Engineering, University of Colorado Boulder 

M.S., Computational and Applied Mathematics, Colorado School of Mines 

B.S., Computational Applied Mathematics and Statistics, Colorado School of Mines

Associations and Memberships

Society for Industrial and Applied Mathematics

Featured Work

Global Sensitivity Metrics from Active SubspacesReliability Engineering & System Safety (2017)

Python Active-subspaces Utility Library Journal of Open Source Software (2016)

Sparse Polynomial Chaos Expansions via Compressed Sensing and D-optimal Design, Computer Methods in Applied Mechanics and Engineering (2018)

A Modified SEIR Model for the Spread of Ebola in Western Africa and Metrics for Resource Allocation, Applied Mathematics and Computation (2018)