Planning to Give Information in Partially Observed Domains with a Learned Weighted Entropy Model

In many robotic applications, an autonomous agent must act within and explore a partially observed environment that is unobserved by its human teammate. We consider such a setting in which the agent can, while acting, transmit declarative information to the human that helps them understand aspects of this unseen environment. Naturally, the human will have preferences about what information they are given. This work adopts an information-theoretic view of the human's preferences: the human scores information based on the induced change in weighted entropy of their belief about the environment state. We formulate this setting as a belief MDP and give an algorithm for solving it approximately. Then, we give an algorithm that allows the agent to learn the human's preferences online. We validate our approach experimentally in simulated discrete and continuous partially observed search-and-recover domains.