Full-text

URL: http://www.cs.york.ac.uk/mlg/MI16/Steels/origins.ps

Related links

CiteSeer: http://citeseer.ist.psu.edu/steels97origin.html

Web search: Google Web Search   ::   Google Scholar

Within this site: References (24)

Paper at a Glance

The Origins of Syntax in Visually Grounded Robotic Agents
Luc Steels
SONY Computer Science Laboratory, 6 Rue Amyot, 75005 Paris
VUB AI Laboratory, Pleinlaan 2, 1050 Brussels
steels@arti.vub.ac.be
In: Pollack, M. (ed.) Proceedings of IJCAI­97.
Morgan Kauffman Publishes, Los Angeles

Abstract The paper proposes a set of principles and a general architecture that may explain how lan­ guage and meaning may originate and com­ plexify in a group of physically grounded dis­ tributed agents. An experimental setup is in­ troduced for concretising and validating spe­ cific mechanisms based on these principles. The setup consists of two robotic heads that watch a scene in which a robot moves around in its ecosystem. The first results from experiments showing the emergence of distinctions, of a lex­ icon, and of primitive syntactic structures are reported.
1 Introduction Artificial Intelligence research has made remarkable progress the last decades by showing how operations over symbolic models may explain various aspects of intelli­ gent behavior, such as planning, problem solving, nat­ ural language processing, etc. However, the problem of the origin of these symbolic models has so far not been adequately addressed. Most of the time it is the programmer who designs formalisms and datastructures, who provides the ontology of objects, concepts and their relations, and who interprets the world and feeds ex­ amples to the AI system. Even most learning systems (including most neural network experiments) start from a prior ontology, carefully designed formalisms or net­ works, and carefully prepared example sets. This gap in current AI has been severely criticised, for example by Searle through his Chinese Room metaphor. The research discussed in this paper attempts to address the lack of grounding and the lack of self­ construction in present­day AI systems. It focuses on how representations could originate and become more complex, without the intervention of human designers. We ...

@inproceedings{steels97theOrigins,
  author={L. Steels},
  title={The origins of syntax in visually grounded robotic agents},
  year={1997},
  address={Los Angeles},
  editor={M. Pollack},
  publisher={Morgan Kauffman Publishers},
  booktitle={IJCAI97},
  url={http://groups.lis.illinois.edu/amag/langev/paper/steels97theOrigins.html}
}