Marginalizing Out Transition Probabilities for Several Subclasses of PFAs

Chihiro Shibata, Ryo Yoshinaka
Proceedings of the Eleventh International Conference on Grammatical Inference, PMLR 21:259-263, 2012.

Abstract

A Bayesian manner which marginalizes transition probabilities can be generally applied to various kinds of probabilistic finite state machine models. Based on such a Bayesian manner, we implemented and compared three algorithms: variable-length gram, state merging method for PDFAs, and collapsed Gibbs sampling for PFAs. Among those, collapsed Gibbs sampling for PFAs performed the best on the data from the pre-competition stage of PAutomaC, although it consumes large computation resources.

Cite this Paper

BibTeX
@InProceedings{pmlr-v21-shibata12a, title = {Marginalizing Out Transition Probabilities for Several Subclasses of PFAs}, author = {Shibata, Chihiro and Yoshinaka, Ryo}, booktitle = {Proceedings of the Eleventh International Conference on Grammatical Inference}, pages = {259--263}, year = {2012}, editor = {Heinz, Jeffrey and Higuera, Colin and Oates, Tim}, volume = {21}, series = {Proceedings of Machine Learning Research}, address = {University of Maryland, College Park, MD, USA}, month = {05--08 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v21/shibata12a/shibata12a.pdf}, url = {https://proceedings.mlr.press/v21/shibata12a.html}, abstract = {A Bayesian manner which marginalizes transition probabilities can be generally applied to various kinds of probabilistic finite state machine models. Based on such a Bayesian manner, we implemented and compared three algorithms: variable-length gram, state merging method for PDFAs, and collapsed Gibbs sampling for PFAs. Among those, collapsed Gibbs sampling for PFAs performed the best on the data from the pre-competition stage of PAutomaC, although it consumes large computation resources.} }
Endnote
%0 Conference Paper %T Marginalizing Out Transition Probabilities for Several Subclasses of PFAs %A Chihiro Shibata %A Ryo Yoshinaka %B Proceedings of the Eleventh International Conference on Grammatical Inference %C Proceedings of Machine Learning Research %D 2012 %E Jeffrey Heinz %E Colin Higuera %E Tim Oates %F pmlr-v21-shibata12a %I PMLR %P 259--263 %U https://proceedings.mlr.press/v21/shibata12a.html %V 21 %X A Bayesian manner which marginalizes transition probabilities can be generally applied to various kinds of probabilistic finite state machine models. Based on such a Bayesian manner, we implemented and compared three algorithms: variable-length gram, state merging method for PDFAs, and collapsed Gibbs sampling for PFAs. Among those, collapsed Gibbs sampling for PFAs performed the best on the data from the pre-competition stage of PAutomaC, although it consumes large computation resources.
RIS
TY - CPAPER TI - Marginalizing Out Transition Probabilities for Several Subclasses of PFAs AU - Chihiro Shibata AU - Ryo Yoshinaka BT - Proceedings of the Eleventh International Conference on Grammatical Inference DA - 2012/08/16 ED - Jeffrey Heinz ED - Colin Higuera ED - Tim Oates ID - pmlr-v21-shibata12a PB - PMLR DP - Proceedings of Machine Learning Research VL - 21 SP - 259 EP - 263 L1 - http://proceedings.mlr.press/v21/shibata12a/shibata12a.pdf UR - https://proceedings.mlr.press/v21/shibata12a.html AB - A Bayesian manner which marginalizes transition probabilities can be generally applied to various kinds of probabilistic finite state machine models. Based on such a Bayesian manner, we implemented and compared three algorithms: variable-length gram, state merging method for PDFAs, and collapsed Gibbs sampling for PFAs. Among those, collapsed Gibbs sampling for PFAs performed the best on the data from the pre-competition stage of PAutomaC, although it consumes large computation resources. ER -
APA
Shibata, C. & Yoshinaka, R.. (2012). Marginalizing Out Transition Probabilities for Several Subclasses of PFAs. Proceedings of the Eleventh International Conference on Grammatical Inference, in Proceedings of Machine Learning Research 21:259-263 Available from https://proceedings.mlr.press/v21/shibata12a.html.

Related Material