<OT> New Posting: ROA-970

[roa at ruccs.rutgers.edu]

ROA 970-0508

Convergence Properties of a Gradual Learning Algorithm for Harmonic Grammar

Paul Boersma <paul.boersma at uva.nl>
Joe Pater <pater at linguist.umass.edu>

Direct link: http://roa.rutgers.edu/view.php3?roa=970


Abstract:
This paper investigates a gradual on-line learning algorithm
for Harmonic Grammar. By adapting existing convergence proofs
for perceptrons, we show 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. We also prove
convergence 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.

Comments: 
Keywords: Harmonic Grammar, Stochastic OT, Noisy HG, perceptron
Areas: Learnability,Computation,Language Acquisition
Type: Manuscript

Direct link: http://roa.rutgers.edu/view.php3?roa=970