Situated Grounded Word Semantics

Abstract

The paper reports on experiments in which autonomous visually grounded agents bootstrap an ontology and a shared lexicon without prior design nor other forms of human intervention. The agents do so while playing a particular language game called the guessing game. We show that synonymy and polysemy arise as emergent properties in the language but also that there are tendencies to dampen it so as to make the language more coherent and thus more optimal from the viewpoints of communicative success, cognitive complexity, and learnability. 1 Introduction The goal of studying natural language semantics is to determine the systematic relations between language utterances, their meanings and their referents. Speakers must conceptualise reality to find an adequate meaning, and they verbalise this meaning to yield an utterance transmitted to the hearer. The hearer must interpret the utterance to reconstruct the meaning and apply the meaning in this particular context to retrieve ...

Cite

Text

Steels and Kaplan. "Situated Grounded Word Semantics." International Joint Conference on Artificial Intelligence, 1999.

Markdown

[Steels and Kaplan. "Situated Grounded Word Semantics." International Joint Conference on Artificial Intelligence, 1999.](https://mlanthology.org/ijcai/1999/steels1999ijcai-situated/)

BibTeX

@inproceedings{steels1999ijcai-situated,
  title     = {{Situated Grounded Word Semantics}},
  author    = {Steels, Luc and Kaplan, Frédéric},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {1999},
  pages     = {862-867},
  url       = {https://mlanthology.org/ijcai/1999/steels1999ijcai-situated/}
}