Meta-Cognition. An Inverse-Inverse Reinforcement Learning Approach for Cognitive Radars

Detecting Changes in Strategy of a Cognitive Radar: a Stochastic Revealed Preference Framework

“…2, might be incorrect. It's possible that the radar is allocating a portion of its beam time to additional targets as well, (iii) If the radar gets to know its being probed, it may intentionally take sub-optimal decisions, to confuse the enemy, as commonly used in ECM's and ECCM's, such as metacognitive radar in [8].…”

“…CR, in general, can be considered as a constrained utility maximizer, and articulating the system goals in a mathematical form suitable for optimization is thus critical to its operation [6]. Revealed preference, a concept originally developed in microeconomics, is a framework for non-parametric learning of utility maximizing behaviour [7], and has been widely applied in the context of inverse learning in cognitive radar [5], [8].…”