Deep Learning for Efficient Discriminative Parsing

Ronan Collobert
Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, PMLR 15:224-232, 2011.

Abstract

We propose a new fast purely discriminative algorithm for natural language parsing, based on a “deep” recurrent convolutional graph transformer network (GTN). Assuming a decomposition of a parse tree into a stack of “levels”, the network predicts a level of the tree taking into account predictions of previous levels. Using only few basic text features which leverage word representations from Collobert and Weston (2008), we show similar performance (in $F_1$ score) to existing pure discriminative parsers and existing “benchmark” parsers (like Collins parser, probabilistic context-free grammars based), with a huge speed advantage.

Cite this Paper


BibTeX
@InProceedings{pmlr-v15-collobert11a, title = {Deep Learning for Efficient Discriminative Parsing}, author = {Collobert, Ronan}, booktitle = {Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics}, pages = {224--232}, year = {2011}, editor = {Gordon, Geoffrey and Dunson, David and Dudík, Miroslav}, volume = {15}, series = {Proceedings of Machine Learning Research}, address = {Fort Lauderdale, FL, USA}, month = {11--13 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v15/collobert11a/collobert11a.pdf}, url = {https://proceedings.mlr.press/v15/collobert11a.html}, abstract = {We propose a new fast purely discriminative algorithm for natural language parsing, based on a “deep” recurrent convolutional graph transformer network (GTN). Assuming a decomposition of a parse tree into a stack of “levels”, the network predicts a level of the tree taking into account predictions of previous levels. Using only few basic text features which leverage word representations from Collobert and Weston (2008), we show similar performance (in $F_1$ score) to existing pure discriminative parsers and existing “benchmark” parsers (like Collins parser, probabilistic context-free grammars based), with a huge speed advantage.} }
Endnote
%0 Conference Paper %T Deep Learning for Efficient Discriminative Parsing %A Ronan Collobert %B Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2011 %E Geoffrey Gordon %E David Dunson %E Miroslav Dudík %F pmlr-v15-collobert11a %I PMLR %P 224--232 %U https://proceedings.mlr.press/v15/collobert11a.html %V 15 %X We propose a new fast purely discriminative algorithm for natural language parsing, based on a “deep” recurrent convolutional graph transformer network (GTN). Assuming a decomposition of a parse tree into a stack of “levels”, the network predicts a level of the tree taking into account predictions of previous levels. Using only few basic text features which leverage word representations from Collobert and Weston (2008), we show similar performance (in $F_1$ score) to existing pure discriminative parsers and existing “benchmark” parsers (like Collins parser, probabilistic context-free grammars based), with a huge speed advantage.
RIS
TY - CPAPER TI - Deep Learning for Efficient Discriminative Parsing AU - Ronan Collobert BT - Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics DA - 2011/06/14 ED - Geoffrey Gordon ED - David Dunson ED - Miroslav Dudík ID - pmlr-v15-collobert11a PB - PMLR DP - Proceedings of Machine Learning Research VL - 15 SP - 224 EP - 232 L1 - http://proceedings.mlr.press/v15/collobert11a/collobert11a.pdf UR - https://proceedings.mlr.press/v15/collobert11a.html AB - We propose a new fast purely discriminative algorithm for natural language parsing, based on a “deep” recurrent convolutional graph transformer network (GTN). Assuming a decomposition of a parse tree into a stack of “levels”, the network predicts a level of the tree taking into account predictions of previous levels. Using only few basic text features which leverage word representations from Collobert and Weston (2008), we show similar performance (in $F_1$ score) to existing pure discriminative parsers and existing “benchmark” parsers (like Collins parser, probabilistic context-free grammars based), with a huge speed advantage. ER -
APA
Collobert, R.. (2011). Deep Learning for Efficient Discriminative Parsing. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 15:224-232 Available from https://proceedings.mlr.press/v15/collobert11a.html.

Related Material