Workshop on Models and Representations for Natural Human-Robot Communication (Robotics: Science and Systems)

Optimal Semantic Distance for Negative Example Selection in Grounded Language Acquisition

, , and

Grounded language acquisition, in which the meanings of utterances are learned from and with respect to the physical world, is often treated as a data-driven machine learning problem. For a robot, obtaining negative examples of language referents is a challenging problem: people tend to describe things that are true of a situation, rather than negatives about it.

  • 3846331 bytes

Downloads: 167 downloads

UMBC ebiquity