We develop fast algorithms for estimation of generalized linear models with convex penalties. The models include linear regression, two-class logistic regression, and multinomial regression problems while the penalties include ℓ1 (the lasso), ℓ2 (ridge regression) and mixtures of the two (the elastic net). The algorithms use cyclical coordinate descent, computed along a regularization path. The methods can handle large problems and can also deal efficiently with sparse features. In comparative timings we find that the new algorithms are considerably faster than competing methods.
%0 Journal Article
%1 friedman_regularization_2010
%A Friedman, Jerome
%A Hastie, Trevor
%A Tibshirani, Rob
%D 2010
%J Journal of statistical software
%K L1, Lasso, Linear Models, optimization, regularization
%N 1
%P 1--22
%T Regularization paths for generalized linear models via coordinate descent
%U http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2929880/
%V 33
%X We develop fast algorithms for estimation of generalized linear models with convex penalties. The models include linear regression, two-class logistic regression, and multinomial regression problems while the penalties include ℓ1 (the lasso), ℓ2 (ridge regression) and mixtures of the two (the elastic net). The algorithms use cyclical coordinate descent, computed along a regularization path. The methods can handle large problems and can also deal efficiently with sparse features. In comparative timings we find that the new algorithms are considerably faster than competing methods.
@article{friedman_regularization_2010,
abstract = {We develop fast algorithms for estimation of generalized linear models with convex penalties. The models include linear regression, two-class logistic regression, and multinomial regression problems while the penalties include ℓ1 (the lasso), ℓ2 (ridge regression) and mixtures of the two (the elastic net). The algorithms use cyclical coordinate descent, computed along a regularization path. The methods can handle large problems and can also deal efficiently with sparse features. In comparative timings we find that the new algorithms are considerably faster than competing methods.},
added-at = {2017-01-09T13:57:26.000+0100},
author = {Friedman, Jerome and Hastie, Trevor and Tibshirani, Rob},
biburl = {https://www.bibsonomy.org/bibtex/2f7538fc3162bbe7ffa355e04aab2433c/yourwelcome},
interhash = {05adad9e58df9e5272eab28621cbc3e8},
intrahash = {f7538fc3162bbe7ffa355e04aab2433c},
issn = {1548-7660},
journal = {Journal of statistical software},
keywords = {L1, Lasso, Linear Models, optimization, regularization},
number = 1,
pages = {1--22},
pmcid = {PMC2929880},
pmid = {20808728},
timestamp = {2017-01-09T14:01:11.000+0100},
title = {Regularization paths for generalized linear models via coordinate descent},
url = {http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2929880/},
urldate = {2013-04-08},
volume = 33,
year = 2010
}