Harmonic Grammar and Harmonic Serialism - John J. McCarthy

Harmonic Grammar and Harmonic Serialism - John J. McCarthy

13. Convergence Properties of a Gradual Learning Algorithm for Harmonic Grammar

Harmonic Grammar and Harmonic Serialism - John J. McCarthy

Paul Boersma [+-]
University of Amsterdam
Paul Boersma is Professor of Phonetics at the University of Amsterdam.
Joe Pater [+-]
University of Massachusetts Amherst
Joe Pater is Professor at the University of Massachusetts Amherst. He received his PhD in 1997 from McGill University, and specializes in phonological theory, phonological acquisition and learning theory

Description

This paper investigates a gradual on-line learning algorithm for Harmonic Grammar. By adapting existing convergence proofs for perceptrons, it shows that for any nonvarying target language, Harmonic-Grammar learners are guaranteed to converge to an appropriate grammar, if they receive complete information about the structure of the learning data. It also prove convergences when the learner incorporates evaluation noise, as in Stochastic Optimality Theory. Computational tests of the algorithm show that it converges quickly. When learners receive incomplete information (e.g. some structure remains hidden), tests indicate that the algorithm is more likely to converge than two comparable Optimality-Theoretic learning algorithms.

Notify A Colleague

Citation

Boersma, Paul; Pater, Joe. 13. Convergence Properties of a Gradual Learning Algorithm for Harmonic Grammar. Harmonic Grammar and Harmonic Serialism. Equinox eBooks Publishing, United Kingdom. p. 389-434 Sep 2016. ISBN 9781845531492. https://www.equinoxpub.com/home/view-chapter/?id=24952. Date accessed: 18 Oct 2018 doi: 10.1558/equinox.24952. Sep 2016

Dublin Core Metadata