# Temporal Graph Neural Network **Entity class:** AI And Graph-Learning Layer concept **Temporal Graph Neural Network** is a representation in which entities or states are nodes and typed relationships or transitions are edges. Used correctly, Temporal Graph Neural Network supports conditional inference and collection planning without promoting proximity, exposure, or model output into proof. ## 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/Graph Machine Learning|Graph Machine Learning]]. - **ontology neighbors:** [[wiki/Graph Transformer|Graph Transformer]] and [[wiki/Dynamic Graph Neural Network|Dynamic Graph Neural Network]]. - **synthesis:** [[wiki/Counterterrorism Predictive Graph|Counterterrorism Predictive Graph]] and [[wiki/Predictive Intelligence Loop|Predictive Intelligence Loop]]. ## Sources - [IARPA — AGILE technical overview (accessed September 23, 2026)](https://www.iarpa.gov/images/research-programs/AGILE/AGILE_ModSim_-_Final.pdf) - [IARPA — Making Data Analysis More AGILE (accessed September 23, 2026)](https://www.iarpa.gov/newsroom/article/making-data-analysis-more-agile) **As of:** 2026-09-23