# Bayesian Inverse Problem **Entity class:** Probabilistic And Inverse-Problem Layer concept **Bayesian Inverse Problem** is a probabilistic construct for updating uncertainty about hidden states or competing explanations when evidence arrives. Bayesian Inverse Problem belongs in the ontology because AI can compute over this structure while analysts retain responsibility for interpretation and authorized action. ## Counterterrorism predictive-graph role The page records capability and analytic function. Any claim of a named operational deployment remains subject to the [[wiki/Counterterrorism Predictive Graph Evidence Ladder|evidence ladder]]. The analytic state should distinguish ground truth, observed evidence, inferred state, and predicted state. ## Relationships - **domain router:** [[wiki/Bayesian Inference|Bayesian Inference]]. - **ontology neighbors:** [[wiki/Bayesian Inference|Bayesian Inference]] and [[wiki/Bayes' Theorem|Bayes' Theorem]]. - **synthesis:** [[wiki/Counterterrorism Predictive Graph|Counterterrorism Predictive Graph]] and [[wiki/Predictive Intelligence Loop|Predictive Intelligence Loop]]. ## Sources - [NIST/SEMATECH — e-Handbook of Statistical Methods (accessed September 23, 2026)](https://www.itl.nist.gov/div898/handbook/) - [Oden Institute — Tan Bui-Thanh research profile (accessed September 23, 2026)](https://oden.utexas.edu/people/directory/Tan-Bui/) **As of:** 2026-09-23