J. González Ortega, D. Ríos Insua, A. Gómez Corral
We consider Differential Games (DGs) corresponding to conflict situations in which players choose strategies over time. In this context, typical applications include defense, counter-terrorism and finance. Attempts to solve DGs have focused mostly on Nash equilibria based methods, but this is not satisfactory in many of the above applications since beliefs and preferences of adversaries will not be readily available, frequently violating game theoretical common knowledge assumptions. Adversarial Risk Analysis (ARA) provides a way forward by supporting one of the players (defender) against the rest (attackers). To do so, ARA minimises the defender's subjective expected costs trying to predict the attackers' (random) optimal actions through the modelling of their decision-making problems under assumptions about their rationality. Our approach is illustrated through a botnet defense example, see Bensoussan et al. (2010), in which we compare both the game theoretical and ARA solutions.
Palabras clave / Keywords: non-cooperative games, decision analysis, nonzero-sum games, cybersecurity
Programado
Sesión V07 Teoría y Procesos de Decisión
1 de junio de 2018 16:00
Sala 6
