• Documents
  • Authors
  • Tables
  • Other Seers ▼
    RefSeer AckSeer CollabSeer SeerSeer
  • Log in
  • Sign up
  • MetaCart

CiteSeerX logo

Advanced Search Include Citations
Advanced Search Include Citations | Disambiguate

Stochastic Attribute-Value Grammars (1997)

Cached

  • Download as a PDF

Download Links

  • [acl.ldc.upenn.edu]
  • [odur.let.rug.nl]
  • [arxiv.org]
  • [www.sfs.nphil.uni-tuebingen.de]
  • [www.vinartus.net]
  • [www.vinartus.com]

  • Other Repositories/Bibliography

  • DBLP
  • Save to List
  • Add to Collection
  • Correct Errors
  • Monitor Changes
by Steven P. Abney
Citations:119 - 0 self
  • Summary
  • Active Bibliography
  • Co-citation
  • Clustered Documents
  • Version History

BibTeX

@MISC{Abney97stochasticattribute-value,
    author = {Steven P. Abney},
    title = {Stochastic Attribute-Value Grammars},
    year = {1997}
}

Bookmark

citeulike Connotea Bibsonomy Del.icio.us Digg Reddit

OpenURL

 

Abstract

Citations

464 Inducing Features of Random Fields - Pietra, Pietra, et al. - 1997
173 Image Analysis, Random Fields and Dynamic Monte Carlo Methods - Winkler - 1995
116 Elementary Principles of Statistical Mechanics - Gibbs - 1902
29 Probabilistic constraint logic programming - RIEZLER - 1998
27 Stochastic HPSG - Brew - 1995
12 Towards probabilistic extensions of constraintbased grammars - Eisele - 1994
12 Parameter estimation for constrained context-free language models - Mark, Miller, et al. - 1992
8 Quantitative constraint logic programming for weighted grammar applications. Talk given at LACL - Riezler - 1996
1 Quantitative constraint logic programming for Abney Stochastic Attribute-Value Grammars weighted grammar applications. Talk given at LACL - Riezler - 1996
1 Weight Estimation In the feature selection step, we choose an initial weight fi for each candidate feature f so as to maximize the gain G = D(~pjjq old ) \Gamma D(~pjjq f;fi ) of adding f to the field. It is actually more convenient to consider log weight - Initial - 1995
1 can re-express this expectation in terms of the old field q old : q ff [f r ] = P x f r (x)q ff (x) = P x f r (x)e fff(x) q old (x) P x e fff(x) q old (x) = q old [f r e fff ] q old [e fff ] The expectations q old [f r e fff ] can be obtained by generatin - We - 1995
1 Entropies, combinatorics and probabilities of context-free branching processes - Miller, O'Sullivan - 1990
1 αt) we require ˜p[f] and qα[f], and F ′ (αt) can be expressed as qα[f] 2 −qα[f 2 ]. ˜p[f] is simply the average value of f in the training corpus. The remaining terms are all of the form qα[f r ]. We can re-express this expectation in terms of the old fie - For - 1995
The National Science Foundation
  • About CiteSeerX
  • Submit Documents
  • Privacy Policy
  • Help
  • Data
  • Source
  • Contact Us

Developed at and hosted by The College of Information Sciences and Technology

© 2007-2010 The Pennsylvania State University