CiteSeerX — Methodological Issues in Simulating the Emergence of Language

Methodological Issues in Simulating the Emergence of Language (2000) [5 citations — 1 self]

Popular Tags

No tags have been applied to this document.

BibTeX | Add To MetaCart

@MISC{Tonkes00methodologicalissues,
    author = {Bradley Tonkes and Janet Wiles},
    title = {Methodological Issues in Simulating the Emergence of Language},
    year = {2000}
}

Bookmarks

OpenURL

Abstract:

this paper we explore Kirby's simulations in greater detail. Kirby credited his results to the `learning bottleneck' but didn't examine variations in learners, tasks or parameters. His choice of learner was motivated only by the fact that it had been developed as an algorithm for grammar induction, and the choice of semantic domain was constrained so as to have combinatorial structure. The question we consider is whether the learning bottleneck is the primary factor with other kinds of learners and a dierently structured domains. In previous work, we have considered communication between a pair of recurrent neural networks. Two networks try to communicate a \concept" represented by a point in the unit interval, [0; 1], over a symbolic channel. One network sends a sequence of symbols for each concept, which the other receives and processes back into a concept. Using this framework, we have shown how a language can evolve to accommodate opposing biases between encoder and decoder [11], and how language evolution can facilitate learning by adapting towards the forms that exploit the weak biases of a general purpose learner [12].

Citations

1140	Finding structure in time – Elman - 1990
321	A learning Algorithm for Continually Running Fully Recurrent Neural Networks – Williams, Zipser - 1989
130	Natural language and natural selection – Pinker, Bloom - 1990
92	ªThe Dynamics of Discrete-Time Computation, with Application to Recurrent Neural Networks and Finite State – Casey - 1996
77	Back propagation is sensitive to initial conditions – Kolen - 1991
49	Innate biases and critical periods: combining evolution and learning in the acquisition of syntax – Batali - 1994
46	Learning, bottlenecks and the evolution of recursive syntax – Kirby - 2002
28	Learning and the emergence of coordinated communication – Oliphant, Batali - 1996
10	Foundations of Recurrent Neural Networks – Siegelmann - 1993
5	A paradox of neural encoders and decoders, or, why don't we talk backwards – Tonkes, Blair, et al. - 1999
1	without natural selection: How compositionality emerges from vocabulary in a population of learners, in The Evolutionary Emergence of Language: Social function and the origins of linguistic – Syntax - 1999