Experiments with Spectral Learning of Latent-Variable PCFGs

Shay B. Cohen, Karl Stratos, Michael Collins, Dean P. Foster, Lyle Ungar

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Latent-variable PCFGs (L-PCFGs) are a highly successful model for natural language parsing. Recent work (Cohen et al., 2012) has introduced a spectral algorithm for parameter estimation of L-PCFGs, which—unlike the EM algorithm—is guaranteed to give consistent parameter estimates (it has PAC-style guarantees of sample complexity). This paper describes experiments using the spectral algorithm. We show that the algorithm provides models with the same accuracy as EM, but is an order of magnitude more efficient. We describe a number of key steps used to obtain this level of performance; these should be relevant to other work on the application of spectral learning algorithms. We view our results as strong empirical evidence for the viability of spectral methods as an alternative to EM.
Original languageEnglish
Title of host publicationProceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Place of PublicationAtlanta, Georgia
PublisherAssociation for Computational Linguistics
Pages148-157
Number of pages10
ISBN (Electronic)978-1-937284-47-3
Publication statusPublished - 1 Jun 2013

Fingerprint

Dive into the research topics of 'Experiments with Spectral Learning of Latent-Variable PCFGs'. Together they form a unique fingerprint.

Cite this