A comparison of algorithms for maximum entropy parameter estimation Export

@inproceedings{citeulike:912392, abstract = {Conditional maximum entropy (ME) models provide a general purpose machine learning technique which has been successfully applied to fields as diverse as computer vision and econometrics, and which is used for a wide variety of classification problems in natural language processing. However, the flexibility of ME models is not without cost. While parameter estimation for ME models is conceptually straightforward, in practice ME models for typical natural language tasks are very large, and may well contain many thousands of free parameters. In this paper, we consider a number of algorithms for estimating the parameters of ME models, including iterative scaling, gradient ascent, conjugate gradient, and variable metric methods. Sur-prisingly, the standardly used iterative scaling algorithms perform quite poorly in comparison to the others, and for all of the test problems, a limited-memory variable metric algorithm outperformed the other choices.}, address = {Morristown, NJ, USA}, author = {Malouf, Robert}, booktitle = {COLING-02: proceedings of the 6th conference on Natural language learning}, citeulike-article-id = {912392}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1118871}, citeulike-linkout-1 = {http://dx.doi.org/10.3115/1118853.1118871}, doi = {10.3115/1118853.1118871}, keywords = {entropy}, pages = {1--7}, posted-at = {2008-12-03 11:16:00}, priority = {2}, publisher = {Association for Computational Linguistics}, title = {A comparison of algorithms for maximum entropy parameter estimation}, url = {http://dx.doi.org/10.3115/1118853.1118871}, year = {2002} }

TY - CONF ID - citeulike:912392 L3 - citeulike-article-id:912392 N2 - Conditional maximum entropy (ME) models provide a general purpose machine learning technique which has been successfully applied to fields as diverse as computer vision and econometrics, and which is used for a wide variety of classification problems in natural language processing. However, the flexibility of ME models is not without cost. While parameter estimation for ME models is conceptually straightforward, in practice ME models for typical natural language tasks are very large, and may well contain many thousands of free parameters. In this paper, we consider a number of algorithms for estimating the parameters of ME models, including iterative scaling, gradient ascent, conjugate gradient, and variable metric methods. Sur-prisingly, the standardly used iterative scaling algorithms perform quite poorly in comparison to the others, and for all of the test problems, a limited-memory variable metric algorithm outperformed the other choices. AD - Morristown, NJ, USA EP - 7 TI - A comparison of algorithms for maximum entropy parameter estimation PB - Association for Computational Linguistics T2 - COLING-02: proceedings of the 6th conference on Natural language learning SP - 1 KW - entropy AU - Malouf, Robert PY - 2002/// UR - http://dx.doi.org/10.3115/1118853.1118871 ER -