Sources — Epistemic Uncertainty
Primary references supporting the historical distinction, uncertainty-quantification treatment, and related formal perspectives associated with epistemic uncertainty.
Foundational Distinction
Review of Recommendations for Probabilistic Seismic Hazard Analysis (1997)
https://www.nationalacademies.org/read/5487
Assessing the Reliability of Complex Models (2012)
https://www.nationalacademies.org/read/13395
Uncertainty Quantification
Sources of Uncertainty and Error — Assessing the Reliability of Complex Models (2012)
https://www.nationalacademies.org/read/13395/chapter/4
Ideas Underlying Quantification of Margins and Uncertainties (QMU): A White Paper (2006)
https://www.sandia.gov/research/publications/details/ideas-underlying-quantification-of-margins-and-uncertaintiesqmu-a-white-pap-2006-09-01/
Related Decision Context
Environmental Decisions in the Face of Uncertainty (2013)
https://www.nationalacademies.org/read/12568
Structural Interpretation
The distinction between epistemic uncertainty and aleatoric uncertainty developed across probability, risk analysis, modeling, and uncertainty-quantification contexts in which limitations of knowledge had to be separated from variability attributed to the represented process.
The sources above treat epistemic uncertainty through incomplete knowledge, uncertain model or parameter states, representation of uncertainty, and the possibility that additional information may alter the uncertainty state.
This reference focuses on those structural properties rather than any specific estimation algorithm, application domain, or operational decision procedure.
Epistemic and aleatoric classifications should not be treated as universally independent of modeling choices. The sources above support a bounded distinction whose application can depend on the represented model, available information, and system boundary.