Goal Inference as Inverse Planning

Abstract

Infants and adults are adept at inferring agents' goals from incomplete or ambiguous sequences of behavior.We propose a framework for goal inference based on inverse planning, in which observers invert a probabilistic generative model of goal-dependent plans to infer agents' goals.The inverse planning framework encompasses many specific models and representations; we present several specific models and test them in two behavioral experiments on online and retrospective goal inference.