Learning Near-Optimal Intrusion Responses Against Dynamic Attackers

Kim Hammar, Rolf Stadler

We study automated intrusion response and formulate the interaction between an attacker and a defender as an optimal stopping game where attack and defense strategies evolve throug…