Blocked Inference in Bayesian Tree Substitution Grammars

12 years 7 months ago
Blocked Inference in Bayesian Tree Substitution Grammars
Learning a tree substitution grammar is very challenging due to derivational ambiguity. Our recent approach used a Bayesian non-parametric model to induce good derivations from treebanked input (Cohn et al., 2009), biasing towards small grammars composed of small generalisable productions. In this paper we present a novel training method for the model using a blocked Metropolis-Hastings sampler in place of the previous method's local Gibbs sampler. The blocked sampler makes considerably larger moves than the local sampler and consequently converges in less time. A core component of the algorithm is a grammar transformation which represents an infinite tree substitution grammar in a finite context free grammar. This enables efficient blocked inference for training and also improves the parsing algorithm. Both algorithms are shown to improve parsing accuracy.
Trevor Cohn, Phil Blunsom
Added 10 Feb 2011
Updated 10 Feb 2011
Type Journal
Year 2010
Where ACL
Authors Trevor Cohn, Phil Blunsom
Comments (0)