Methods of Moments for Learning Stochastic Languages: Unified Presentation and Empirical Comparison

Borja Balle, William Hamilton, Joelle Pineau
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):1386-1394, 2014.

Abstract

Probabilistic latent-variable models are a powerful tool for modelling structured data. However, traditional expectation-maximization methods of learning such models are both computationally expensive and prone to local-minima. In contrast to these traditional methods, recently developed learning algorithms based upon the method of moments are both computationally efficient and provide strong statistical guarantees. In this work, we provide a unified presentation and empirical comparison of three general moment-based methods in the context of modelling stochastic languages. By rephrasing these methods upon a common theoretical ground, introducing novel theoretical results where necessary, we provide a clear comparison, making explicit the statistical assumptions upon which each method relies. With this theoretical grounding, we then provide an in-depth empirical analysis of the methods on both real and synthetic data with the goal of elucidating performance trends and highlighting important implementation details.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-balle14, title = {Methods of Moments for Learning Stochastic Languages: Unified Presentation and Empirical Comparison}, author = {Balle, Borja and Hamilton, William and Pineau, Joelle}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {1386--1394}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/balle14.pdf}, url = {https://proceedings.mlr.press/v32/balle14.html}, abstract = {Probabilistic latent-variable models are a powerful tool for modelling structured data. However, traditional expectation-maximization methods of learning such models are both computationally expensive and prone to local-minima. In contrast to these traditional methods, recently developed learning algorithms based upon the method of moments are both computationally efficient and provide strong statistical guarantees. In this work, we provide a unified presentation and empirical comparison of three general moment-based methods in the context of modelling stochastic languages. By rephrasing these methods upon a common theoretical ground, introducing novel theoretical results where necessary, we provide a clear comparison, making explicit the statistical assumptions upon which each method relies. With this theoretical grounding, we then provide an in-depth empirical analysis of the methods on both real and synthetic data with the goal of elucidating performance trends and highlighting important implementation details.} }
Endnote
%0 Conference Paper %T Methods of Moments for Learning Stochastic Languages: Unified Presentation and Empirical Comparison %A Borja Balle %A William Hamilton %A Joelle Pineau %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-balle14 %I PMLR %P 1386--1394 %U https://proceedings.mlr.press/v32/balle14.html %V 32 %N 2 %X Probabilistic latent-variable models are a powerful tool for modelling structured data. However, traditional expectation-maximization methods of learning such models are both computationally expensive and prone to local-minima. In contrast to these traditional methods, recently developed learning algorithms based upon the method of moments are both computationally efficient and provide strong statistical guarantees. In this work, we provide a unified presentation and empirical comparison of three general moment-based methods in the context of modelling stochastic languages. By rephrasing these methods upon a common theoretical ground, introducing novel theoretical results where necessary, we provide a clear comparison, making explicit the statistical assumptions upon which each method relies. With this theoretical grounding, we then provide an in-depth empirical analysis of the methods on both real and synthetic data with the goal of elucidating performance trends and highlighting important implementation details.
RIS
TY - CPAPER TI - Methods of Moments for Learning Stochastic Languages: Unified Presentation and Empirical Comparison AU - Borja Balle AU - William Hamilton AU - Joelle Pineau BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-balle14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 1386 EP - 1394 L1 - http://proceedings.mlr.press/v32/balle14.pdf UR - https://proceedings.mlr.press/v32/balle14.html AB - Probabilistic latent-variable models are a powerful tool for modelling structured data. However, traditional expectation-maximization methods of learning such models are both computationally expensive and prone to local-minima. In contrast to these traditional methods, recently developed learning algorithms based upon the method of moments are both computationally efficient and provide strong statistical guarantees. In this work, we provide a unified presentation and empirical comparison of three general moment-based methods in the context of modelling stochastic languages. By rephrasing these methods upon a common theoretical ground, introducing novel theoretical results where necessary, we provide a clear comparison, making explicit the statistical assumptions upon which each method relies. With this theoretical grounding, we then provide an in-depth empirical analysis of the methods on both real and synthetic data with the goal of elucidating performance trends and highlighting important implementation details. ER -
APA
Balle, B., Hamilton, W. & Pineau, J.. (2014). Methods of Moments for Learning Stochastic Languages: Unified Presentation and Empirical Comparison. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):1386-1394 Available from https://proceedings.mlr.press/v32/balle14.html.

Related Material