Uncertainty Quantification (UQ)
Definition
Uncertainty quantification makes the uncertainty of models and results measurable.
The quantitative description, propagation, and evaluation of uncertainty in inputs, models, and results.
Simply explained
Instead of just giving a single number, a prediction also shows how uncertain it is—for example, through a possible value range.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
A forecasting system provides demand intervals and checks their empirical coverage per product group.
Limits and pitfalls
- A single confidence score does not capture all uncertainties; assumptions, calibration, and distribution shifts must be disclosed.
Origin & history
UQ evolved from statistics, metrology, and numerical simulation and was transferred to machine learning.
Use in mid-sized companies
- Possible application: A forecasting system provides demand intervals and checks their empirical coverage per product group.
Frequently asked questions
What does Uncertainty Quantification mean?
The quantitative description, propagation, and evaluation of uncertainty in inputs, models, and results. What should be watched for? A single confidence score does not capture all uncertainties; assumptions, calibration, and distribution shifts must be disclosed.
Sources
- Uncertainty Quantification · NIST, Uncertainty Quantification
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: