Algorithmic Tools for Adversarial Games

This project will develop and validate a Game-Theoretic Terrorist Modus Operandi Detection System (GT-TMODS) for asymmetric threat detection against politically, socially, and culturally diverse adversaries. GT-TMODS will integrate state-of-the-art game theory and simulation research with proven, powerful graph-matching and social-network analysis methods to model and detect asymmetric threat activities through multi-agent adversarial games. The Phase I effort will (i) develop game-theoretic models for asymmetric threats, (ii) perform game-theoretic simulations to discover asymmetric threat patterns and strategies, and (iii) implement and assess a GT-TMODS prototype. The Phase I GT-TMODS Prototype will demonstrate key innovations including: (i) game-theoretic models of the asymmetric threat, (ii) empirical game-theoretic analysis of the asymmetric threat, (iii) approximate solutions for intractable games identifying plausible threat strategies and profiles, and (iv) graph-based threat patterns and SNA signatures derived from the game-theoretic simulation and analysis. GT-TMODS will provide sophisticated technology to a wide audience of potential users in the military and intelligence communities for enhanced national security against asymmetric threats resulting in significant improvements in our response capabilities against politically, socially, and culturally diverse adversaries. GT-TMODS will have a wide range of important defense and commercial applications including terrorist threat detection, industrial espionage detection, financial fraud detection, and business intelligence.