Contentful mental states for robot baby
Paul R. Cohen, Tim Oates, Carole R. Beal, Niall M. Adams · 2002
In this paper we claim that meaningful representations can be learned by programs, although today they are almost al-ways designed by skilled engineers. We discuss several kinds of meaning that representations might have, and focus on a functional notion of meaning as appropriate for programs to learn. Specifically, a representation is meaningful if it incor-porates an indicator of external conditions and if the indica-tor relation informs action. We survey methods for inducing kinds of representations we call structural abstractions. Pro-totypes of sensory time series are one kind of structural ab-straction, and though they are not denoting or compositional, they do support planning. Deictic representations of objects and prototype representations of words enable a program to learn the denotational meanings of words. Finally, we discuss two algorithms designed to find the macroscopic structure of episodes in a domain-independent way.