Learning Context-Free Grammars with a Simplicity Bias

10 years 2 months ago
Learning Context-Free Grammars with a Simplicity Bias
We examine the role of simplicity in directing the induction of context-free grammars from sample sentences. We present a rational reconstruction of Wol 's SNPR { the Gridssystem { which incorporates a bias toward grammars that minimize description length. The algorithm alternates between merging existing nonterminal symbols and creating new symbols, using a beam search to move from complex to simpler grammars. Experiments suggest that this approach can induce accurate grammars and that it scales reasonably to more di cult domains.
Pat Langley, Sean Stromsten
Added 02 Aug 2010
Updated 02 Aug 2010
Type Conference
Year 2000
Where ECML
Authors Pat Langley, Sean Stromsten
Comments (0)