Paper: Products of Random Latent Variable Grammars

ACL ID N10-1003
Title Products of Random Latent Variable Grammars
Venue Human Language Technologies
Session Main Conference
Year 2010

We show that the automatically induced latent variable grammars of Petrov et al. (2006) vary widely in their underlying representations, de- pending on their EM initialization point. We use this to our advantage, combining multiple automatically learned grammars into an un- weighted product model, which gives signif- icantly improved performance over state-of- the-art individual grammars. In our model, the probability of a constituent is estimated as a product of posteriors obtained from multi- ple grammars that differ only in the random seed used for initialization, without any learn- ing or tuning of combination weights. Despite its simplicity, a product of eight automatically learned grammars improves parsing accuracy from 90.2% to 91.8% on English, and from 80.3% to 84.5% on German.