The Acquisition of Linguistic Competence for Communicating Propositional Logic Sentences
We describe some experiments which show how a language expressive enough to allow the communication of meanings of the same complexity as propositional logic formulas can emerge in a population of autonomous agents which have no prior linguistic knowledge. We take an approach based on general purpose cognitive capacities, such as invention, adoption and induction, and on self-organisation principles applied to a particular type of linguistic interaction known as a language game.
These experiments extend previous work by considering a larger population and a much larger search space of grammar rules. In particular the agents are allowed to order the expressions associated with the constituents of a logical formula in arbitrary order in the sentence. Previous work assumed that the expressions associated with the connectives should be always placed in the first position of the sentence. Another difference is that communication is considered successful in a language game if the meaning interpreted by the hearer is logically equivalent to the meaning the speaker had in mind. In previous experiments the meanings of speaker and hearer were required to be syntactically equal. This allows us to observe how a less strict grammar in terms of word order emerges through the self-organisation process, which minimizes the learning effort of the agents by imposing only those order relations among the components of a sentence that are necessary for language understanding.
Unable to display preview. Download preview PDF.
- 2.McCarthy, J.: Formalizing Common Sense. Papers by John McCarthy. Ablex. Edited by Vladimir Lifschitz (1990)Google Scholar
- 3.Piaget, J.: The Equilibration of Cognitive Structures: the Central Problem of Intellectual Development. University of Chicago Press, Chicago (1985)Google Scholar
- 4.Santibáñez, J.: Relación del rendimiento escolar en las áreas de lectura y escritura con las aptitudes mentales y el desarrollo visomotor. Universidad Nacional de Educación a Distancia, D.L., Madrid (1984) ISBN 84-398-2486-6Google Scholar
- 5.Santibáñez, J.: Variables psicopedagógicas relacionadas con el rendimiento en E.G.B. Instituto de Estudios Riojanos, ISBN 84-87252-00-1, Logroño (1988)Google Scholar
- 6.Santibáñez, J.: La evaluación de la escritura: test de escritura para el ciclo inicial, T.E.C.I. CEPE, D.L. Madrid (1989) ISBN 84-86235-87-1Google Scholar
- 7.Sierra, J.: Grounded models as a basis for intuitive reasoning. In: Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence, pp. 401–406 (2001)Google Scholar
- 8.Sierra, J.: Grounded models as a basis for intuitive and deductive reasoning: The acquisition of logical categories. In: Proceedings of the European Conference on Artificial Intelligence, pp. 93–97 (2002)Google Scholar
- 9.Sierra, J.: Grounded models as a basis for intuitive reasoning: the origins of logical categories. In: Papers from AAAI–2001 Fall Symposium on Anchoring Symbols to Sensor Data in Single and Multiple Robot Systems. Technical Report FS-01-01, pp. 101–108. AAAI Press, Menlo Park (2001)Google Scholar
- 10.Steels, L.: The Talking Heads Experiment. Words and Meanings, vol. 1. LABORATORIUM, Antwerpen (1999) (Special Pre-edition)Google Scholar
- 11.Steels, L., Kaplan, F., McIntyre, A., V Looveren, J.: Crucial factors in the origins of word-meaning. In: The Transition to Language, pp. 252–271. Oxford University Press, Oxford (2002)Google Scholar
- 13.Steels, L.: The emergence of grammar in communicating autonomous robotic agents. In: Proceedings of the European Conference on Artificial Intelligence, pp. 764–769. IOS Publishing, Amsterdam (2000)Google Scholar
- 14.Kirby, S.: Learning, bottlenecks and the evolution of recursive syntax. In: Linguistic Evolution through Language Acquisition: Formal and Computational Models, pp. 96–109. Cambridge University Press, Cambridge (2002)Google Scholar
- 15.Stolcke, A.: Bayesian Learning of Probabilistic Language Models. PhD thesis, Univ. of California at Berkeley (1994)Google Scholar
- 17.Steels, L.: The synthetic modeling of language origins. Evolution of Communication 1(1), 1–35 (1997)Google Scholar
- 18.Batali, J.: The negotiation and acquisition of recursive grammars as a result of competition among exemplars. In: Linguistic Evolution through Language Acquisition: Formal and Computational Models, pp. 111–172. Cambridge U.P, Cambridge (2002)Google Scholar
- 19.Steels, L.: Constructivist development of grounded construction grammars. In: Proc. Annual Meeting of Association for Computational Linguistics, pp. 9–16 (2004)Google Scholar
- 21.Hurford, J.: Social transmission favors linguistic generalization. In: The Evolutionary Emergence of Language: Social Function and the Origins of Linguistic Form, pp. 324–352. Cambridge University Press, Cambridge (2000)Google Scholar