Statistical and Computational Tradeoffs in Stochastic Composite Likelihood

Joshua Dillon, Guy Lebanon
; Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, PMLR 5:129-136, 2009.

Abstract

Maximum likelihood estimators are often of limited practical use due to the intensive computation they require. We propose a family of alternative estimators that maximize a stochastic variation of the composite likelihood function. We prove the consistency of the estimators, provide formulas for their asymptotic variance and computational complexity, and discuss experimental results in the context of Boltzmann machines and conditional random fields. The theoretical and experimental studies demonstrate the effectiveness of the estimators in achieving a predefined balance between computational complexity and statistical accuracy.

Cite this Paper


BibTeX
@InProceedings{pmlr-v5-dillon09a, title = {Statistical and Computational Tradeoffs in Stochastic Composite Likelihood}, author = {Joshua Dillon and Guy Lebanon}, booktitle = {Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics}, pages = {129--136}, year = {2009}, editor = {David van Dyk and Max Welling}, volume = {5}, series = {Proceedings of Machine Learning Research}, address = {Hilton Clearwater Beach Resort, Clearwater Beach, Florida USA}, month = {16--18 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v5/dillon09a/dillon09a.pdf}, url = {http://proceedings.mlr.press/v5/dillon09a.html}, abstract = {Maximum likelihood estimators are often of limited practical use due to the intensive computation they require. We propose a family of alternative estimators that maximize a stochastic variation of the composite likelihood function. We prove the consistency of the estimators, provide formulas for their asymptotic variance and computational complexity, and discuss experimental results in the context of Boltzmann machines and conditional random fields. The theoretical and experimental studies demonstrate the effectiveness of the estimators in achieving a predefined balance between computational complexity and statistical accuracy.} }
Endnote
%0 Conference Paper %T Statistical and Computational Tradeoffs in Stochastic Composite Likelihood %A Joshua Dillon %A Guy Lebanon %B Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2009 %E David van Dyk %E Max Welling %F pmlr-v5-dillon09a %I PMLR %J Proceedings of Machine Learning Research %P 129--136 %U http://proceedings.mlr.press %V 5 %W PMLR %X Maximum likelihood estimators are often of limited practical use due to the intensive computation they require. We propose a family of alternative estimators that maximize a stochastic variation of the composite likelihood function. We prove the consistency of the estimators, provide formulas for their asymptotic variance and computational complexity, and discuss experimental results in the context of Boltzmann machines and conditional random fields. The theoretical and experimental studies demonstrate the effectiveness of the estimators in achieving a predefined balance between computational complexity and statistical accuracy.
RIS
TY - CPAPER TI - Statistical and Computational Tradeoffs in Stochastic Composite Likelihood AU - Joshua Dillon AU - Guy Lebanon BT - Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics PY - 2009/04/15 DA - 2009/04/15 ED - David van Dyk ED - Max Welling ID - pmlr-v5-dillon09a PB - PMLR SP - 129 DP - PMLR EP - 136 L1 - http://proceedings.mlr.press/v5/dillon09a/dillon09a.pdf UR - http://proceedings.mlr.press/v5/dillon09a.html AB - Maximum likelihood estimators are often of limited practical use due to the intensive computation they require. We propose a family of alternative estimators that maximize a stochastic variation of the composite likelihood function. We prove the consistency of the estimators, provide formulas for their asymptotic variance and computational complexity, and discuss experimental results in the context of Boltzmann machines and conditional random fields. The theoretical and experimental studies demonstrate the effectiveness of the estimators in achieving a predefined balance between computational complexity and statistical accuracy. ER -
APA
Dillon, J. & Lebanon, G.. (2009). Statistical and Computational Tradeoffs in Stochastic Composite Likelihood. Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, in PMLR 5:129-136

Related Material