Optimally Deceiving a Learning Leader in Stackelberg Games
This solves a general version of a previously limited problem in game theory and machine learning, with implications for security and strategic interactions in AI systems.
The paper addresses the problem of a follower deceiving a learning leader in Stackelberg games by manipulating payoff inputs, showing that it is always possible to compute near-optimal deceptive payoffs across various interaction scenarios.
Recent results in the ML community have revealed that learning algorithms used to compute the optimal strategy for the leader to commit to in a Stackelberg game, are susceptible to manipulation by the follower. Such a learning algorithm operates by querying the best responses or the payoffs of the follower, who consequently can deceive the algorithm by responding as if his payoffs were much different than what they actually are. For this strategic behavior to be successful, the main challenge faced by the follower is to pinpoint the payoffs that would make the learning algorithm compute a commitment so that best responding to it maximizes the follower's utility, according to his true payoffs. While this problem has been considered before, the related literature only focused on the simplified scenario in which the payoff space is finite, thus leaving the general version of the problem unanswered. In this paper, we fill in this gap, by showing that it is always possible for the follower to compute (near-)optimal payoffs for various scenarios about the learning interaction between leader and follower.