HOME   ::  Edited Book List   ::   Book Chapter
Steels, L. and Kaplan, F. (2002) Bootstrapping grounded word semantics. In Ted Briscoe, editor, Linguistic Evolution through Language Acquisition: Formal and Computational Models. Cambridge University Press.
Bookmark:  

Full-text
   URL: http://arti.vub.ac.be/steels/ground.ps.gz
   Cached: PDF-265K    PS-1276K    PS.gz-163K   
   SAVE AS an easy-to-recall long filename:
      Filename format: author--year--title   PDF-265K    PS-163K    :: About GZip'd PS
      Filename format: author--year--title--journal|proceedings|...--pages   PDF-265K    PS-163K   

Related links
   Source: http://www.csl.sony.fr/General/Publicati...ference=steels%3A99b
   CiteSeer: http://citeseer.ist.psu.edu/steels99bootstrapping.html
  Web search: Google Web Search   ::   Google Scholar
  Within this site: Cited by (21)    References (19)

Paper at a Glance 

Bootstrapping Grounded Word Semantics
Luc Steels (1,2) and Frederic Kaplan (1)
(1) Sony Computer Science Laboratory
6 Rue Amyot, 75005 Paris
(2) VUB AI Laboratory
Pleinlaan 2, 1050 Brussels
steels@arti.vub.ac.be
Abstract The paper reports on experiments with a population of visually grounded robotic agents capable of bootstrapping their own ontology and shared lexicon without prior design nor other forms of human in­ tervention. The agents do so while playing a particular language game called the guessing game. We show that synonymy and ambiguity arise as emergent properties in the lexicon, due to the situated grounded character of the agent­environment interaction, but that there are also tendencies to dampen them so as to make the language more coherent and thus more optimal from the viewpoints of communicative success, cognitive complexity, and learnability.
1 Introduction How do words get their meanings? An answer to this question requires a the­ ory on the origins of meanings, a theory how forms get recruited for express­ ing meanings, and a theory how associations between forms and meanings may propagate in a population. Each theory must characterise properties of a cognitive agent's architecture: components a cognitive agent needs to have, and details on how the different components coordinate their activities. More specifically, the theories should detail what kind of associative memory the agents must have for storing and acquiring form­meaning relations, what type of mechanisms they might use to categorise the environment through sensory inputs, how they might acquire a repertoire of perceptually grounded categories (an ontology), and what behaviors the agents must be capable of so as to communicate successfully through language. To allow validation, theories of agent architecture should be formally specified and hence testable through computer simulations or even better through experiments with robotic agents interacting with real world envi­ ronment ...
BibTex
@incollection{steels02bootstrappingGrounded,
  author={L. Steels and F. Kaplan},
  title={Bootstrapping grounded word semantics},
  year={2002},
  chapter={3},
  editor={Ted Briscoe},
  publisher={Cambridge University Press},
  booktitle={Linguistic Evolution through Language Acquisition: Formal and Computational Models},
  url={http://www.isrl.uiuc.edu/~amag/langev/paper/steels02bootstrappingGrounded.html},
  keywords={semantics, grounding, evolutionary linguistics,Talking Heads}
}

 HOME   ::  Edited Book List   ::   Book Chapter Comments to: junwang4 you-know-at gmail.com Last update: 2/2/08