Parsing as an energy minimization problem. Selman, B. & Hirst, G. Adriaens, G. & Hahn, U., editors. Parallel natural language processing, pages 238–254. Ablex Publishing, Norwood, NJ, 1994. abstract bibtex We show how parsing can be formulated as an energy minimization problem. We describe our model in terms of a connectionist network. In this network we use a parallel computation scheme similar to the one used in the Boltzmann machine and apply simulated annealing, a stochastic optimization method. We show that at low temperatures the time average of the visited states at thermal equilibrium represents the correct parse of the input sentence.
Our treatment of parsing as an energy minimization problem enables us to formulate general rules for the setting of weights and thresholds in the network. We also consider an alternative to the commonly used energy function. Using this new function, one can choose a provably correct set of weights and thresholds for the network and still have an acceptable rate of convergence in the annealing process.
The parsing scheme is built from a small set of connectionist primitives that represent the grammar rules. These primitives are linked together using pairs of computing units that behave like discrete switches. These units are used as binders between concepts represented in the network. They can be linked in such a way that individual rules can be selected from a collection of rules, and are very useful in the construction of connectionist schemes for any form of rule-based processing.
@InBook{ selman1,
author = {Bart Selman and Graeme Hirst},
chapter = {Parsing as an energy minimization problem},
editor = {Geert Adriaens and Udo Hahn},
title = {Parallel natural language processing},
address = {Norwood, NJ},
publisher = {Ablex Publishing},
year = {1994},
pages = {238--254},
abstract = {<P> We show how parsing can be formulated as an energy
minimization problem. We describe our model in terms of a
connectionist network. In this network we use a parallel
computation scheme similar to the one used in the Boltzmann
machine and apply simulated annealing, a stochastic
optimization method. We show that at low temperatures the
time average of the visited states at thermal equilibrium
represents the correct parse of the input sentence.</p> <P>
Our treatment of parsing as an energy minimization problem
enables us to formulate general rules for the setting of
weights and thresholds in the network. We also consider an
alternative to the commonly used energy function. Using
this new function, one can choose a provably correct set of
weights and thresholds for the network and still have an
acceptable rate of convergence in the annealing
process.</p> <P> The parsing scheme is built from a small
set of <I>connectionist primitives</I> that represent the
grammar rules. These primitives are linked together using
pairs of computing units that behave like discrete
switches. These units are used as binders between concepts
represented in the network. They can be linked in such a
way that individual rules can be selected from a collection
of rules, and are very useful in the construction of
connectionist schemes for any form of rule-based
processing.</p>}
}
Downloads: 0
{"_id":{"_str":"521afb59aa2f288d1f000b45"},"__v":2,"authorIDs":[],"author_short":["Selman, B.","Hirst, G."],"bibbaseid":"selman-hirst-parallelnaturallanguageprocessing-1994","bibdata":{"bibtype":"inbook","type":"inbook","author":[{"firstnames":["Bart"],"propositions":[],"lastnames":["Selman"],"suffixes":[]},{"firstnames":["Graeme"],"propositions":[],"lastnames":["Hirst"],"suffixes":[]}],"chapter":"Parsing as an energy minimization problem","editor":[{"firstnames":["Geert"],"propositions":[],"lastnames":["Adriaens"],"suffixes":[]},{"firstnames":["Udo"],"propositions":[],"lastnames":["Hahn"],"suffixes":[]}],"title":"Parallel natural language processing","address":"Norwood, NJ","publisher":"Ablex Publishing","year":"1994","pages":"238–254","abstract":"<P> We show how parsing can be formulated as an energy minimization problem. We describe our model in terms of a connectionist network. In this network we use a parallel computation scheme similar to the one used in the Boltzmann machine and apply simulated annealing, a stochastic optimization method. We show that at low temperatures the time average of the visited states at thermal equilibrium represents the correct parse of the input sentence.</p> <P> Our treatment of parsing as an energy minimization problem enables us to formulate general rules for the setting of weights and thresholds in the network. We also consider an alternative to the commonly used energy function. Using this new function, one can choose a provably correct set of weights and thresholds for the network and still have an acceptable rate of convergence in the annealing process.</p> <P> The parsing scheme is built from a small set of <I>connectionist primitives</I> that represent the grammar rules. These primitives are linked together using pairs of computing units that behave like discrete switches. These units are used as binders between concepts represented in the network. They can be linked in such a way that individual rules can be selected from a collection of rules, and are very useful in the construction of connectionist schemes for any form of rule-based processing.</p>","bibtex":"@InBook{\t selman1,\n author\t= {Bart Selman and Graeme Hirst},\n chapter\t= {Parsing as an energy minimization problem},\n editor\t= {Geert Adriaens and Udo Hahn},\n title\t\t= {Parallel natural language processing},\n address\t= {Norwood, NJ},\n publisher\t= {Ablex Publishing},\n year\t\t= {1994},\n pages\t\t= {238--254},\n abstract\t= {<P> We show how parsing can be formulated as an energy\n\t\t minimization problem. We describe our model in terms of a\n\t\t connectionist network. In this network we use a parallel\n\t\t computation scheme similar to the one used in the Boltzmann\n\t\t machine and apply simulated annealing, a stochastic\n\t\t optimization method. We show that at low temperatures the\n\t\t time average of the visited states at thermal equilibrium\n\t\t represents the correct parse of the input sentence.</p> <P>\n\t\t Our treatment of parsing as an energy minimization problem\n\t\t enables us to formulate general rules for the setting of\n\t\t weights and thresholds in the network. We also consider an\n\t\t alternative to the commonly used energy function. Using\n\t\t this new function, one can choose a provably correct set of\n\t\t weights and thresholds for the network and still have an\n\t\t acceptable rate of convergence in the annealing\n\t\t process.</p> <P> The parsing scheme is built from a small\n\t\t set of <I>connectionist primitives</I> that represent the\n\t\t grammar rules. These primitives are linked together using\n\t\t pairs of computing units that behave like discrete\n\t\t switches. These units are used as binders between concepts\n\t\t represented in the network. They can be linked in such a\n\t\t way that individual rules can be selected from a collection\n\t\t of rules, and are very useful in the construction of\n\t\t connectionist schemes for any form of rule-based\n\t\t processing.</p>}\n}\n\n","author_short":["Selman, B.","Hirst, G."],"editor_short":["Adriaens, G.","Hahn, U."],"key":"selman1","id":"selman1","bibbaseid":"selman-hirst-parallelnaturallanguageprocessing-1994","role":"author","urls":{},"metadata":{"authorlinks":{}}},"bibtype":"inbook","biburl":"www.cs.toronto.edu/~fritz/tmp/compling.bib","downloads":0,"keywords":[],"search_terms":["parallel","natural","language","processing","selman","hirst"],"title":"Parallel natural language processing","title_words":["parallel","natural","language","processing"],"year":1994,"dataSources":["n8jB5BJxaeSmH6mtR","6b6A9kbkw4CsEGnRX"]}