Interdependent defense games: modeling interdependent security under deliberate attacks

摘要: We propose interdependent defense (IDD) games, a computational game-theoretic framework to study aspects of the interdependence risk and security in multi-agent systems under deliberate external attacks. Our model builds upon (IDS) due Heal Kunreuther that considers source be result fixed randomized-strategy. adapt IDS games attacker's behavior. define pure-strategy space utility function derive appropriate cost functions for defenders. provide complete characterization mixed-strategy Nash equilibria (MSNE), design simple polynomial-time algorithm computing all them, an important subclass IDD games. In addition, we random-instance generator (general) based on version real-world Internet-derived Autonomous Systems (AS) graph (with around 27K nodes 100K edges), present promising empirical results using learning heuristics compute (approximate) MSNE such