L1‐regularization path algorithm for generalized linear models

By Mee Young Park, Trevor J. Hastie. 2007

883 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering, Statistics and Probability, Statistical Learning Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

In Statistics and Probability

In Statistical Learning Theory

Citations in the corpus

Learning to Solve Arithmetic Word Problems with Verb Categorization (2014, background): “Solving the problem consists of two main steps: (1) progressing states based on verb categories in sentences and (2) forming the equation.”
Regularization Paths for Generalized Linear Models via Coordinate Descent (2010, background): “`1 regularization paths for generalized linear models [Park and Hastie, 2006]; 5.”
Nearly unbiased variable selection under minimax concave penalty (2010, neutral): “…main difference of such an extension from the computation of the LASSO for the generalized linear models [Genkin, Lewis and Madigan (2004), Zhao and Yu (2007) and Park and Hastie (2007)] is the possibility of the sign change dλ(x)/dx to allow the path to traverse from one local minimum to another.”
A SIGNIFICANCE TEST FOR THE LASSO (2013, methodology): “However, one can numerically approximate these knot values; for example, see Park and Hastie (2007). By analogy to the Gaussian case, we would hope that Tk has an asymptotic Exp(1) distribution under the null. Though we have not rigorously investigated this conjecture, simulations seem to support it”
ADAPTIVE SEMI-VARYING COEFFICIENT MODEL SELECTION (2012, background): “These include the shooting algorithm (Fu (1998); Yuan and Lin (2006)), local quadratic approximation (Fan and Li (2001)), least angle regression (Efron et al. (2004)), and many others (Zhao and Yu (2004); Park and Hastie (2007); Zou and Li (2008)).”
High-Dimensional Variable Selection for Survival Data (2010, methodology): “We tested RSF-VH on five different benchmark microarray data sets: the diffuse large B-cell lymphoma (DLBCL) data set of Rosenwald et al. (2002), the breast cancer data set of van’t Veer et al.”
Penalized regression, standard errors, and Bayesian lassos (2010, methodology): “3 The Bayesian Lasso Tibshirani (1996) noted that with the L1 penalty term in (2), the lasso estimates could be interpreted as the Bayes posterior mode under independent Laplace (doubleexponential) priors for the βjs.”
Learning Graphical Model Structure Using L1-Regularization Paths (2007, background): “However, predictor-corrector strategies can be used for following such curved regularization paths (see ( PH06 )).”
Boosting Algorithms: Regularization, Prediction and Model Fitting. Rejoinder. (2007, methodology): “The approximate equivalence is derived by representing FSLR and Lasso as two different modifications of the computationally efficient least angle regression (LARS) algorithm from Efron et al. (author?) [28] (see also [ 68 ] for generalized linear models).”