Reproducible Research, Uncertainty Quantification, and Verification & Validation
Slides used with my presentation in the SIAM Uncertainty Quantification Conference 2014, Minisymposium on "The Reliability of Computational Research Findings: Reproducible Research, Uncertainty Quantification, and Verification & Validation."
The talk used an audience response system to collect True/False or Yes/No opinions on 13 statements/questions:
1) Computer simulations create scientific knowledge.
2) Simulation is a method
3) A reproducible simulation does not need to be accurate.
4) Is there value to a Reproducible Wrong Answer?
5) Simulation is an experiment.
6) Is V&V a pre-requisite to Reproducibility?
7) UQ aims to give objective confidence levels for the results of simulations.
8) Is V&V a pre-requisite to UQ?
9) UQ presupposes verification and informs validation.
10) Verification should be done before validation.
11) In verification, the actual value of the error is generally unknown.
12) Can reproducible science be uncertain?
13) UQ is a requirement of reproducibility.
The slide deck was here augmented with the screenshots from the audience response.