Answers are grounded in this paper and cite their sources.
Autonomous systems are increasingly expected to operate in the presence of adversaries, though adversaries may infer sensitive information simply by observing a system. Therefore, we present a deceptive sequential decision-making framework that not only conceals sensitive information, but actively misleads adversaries about it. We model autonomous systems as Markov decision processes, with adversaries using inverse reinforcement learning to recover reward functions. To counter them, we present three regularisation strategies for policy synthesis problems that actively deceive an adversary about a system’s reward. “Diversionary deception” leads an adversary to draw any false conclusion about a system’s reward function. “Targeted deception” leads an adversary to draw a specific false conclusion about a system’s reward function. “Equivocal deception” leads an adversary to infer that the real reward and a false reward both explain a system’s behaviour. We show how each form of deception can be implemented in policy optimisation problems and analytically bound the loss in total accumulated reward induced by deception. Next, we empirically evaluate these developments. We show that diversionary, targeted, and equivocal deception all steer the adversary to false beliefs while still attaining a total accumulated reward that is approximately 85% to 99% of its optimal, non-deceptive value.
No takes yet. Share an insight, caveat, or question.
Kim et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: