Strong Rules for Discarding Predictors in Lasso-Type Problems
From MaRDI portal
Publication:5743136
DOI10.1111/J.1467-9868.2011.01004.XzbMATH Open1411.62213arXiv1011.2234OpenAlexW2131060185WikidataQ34674446 ScholiaQ34674446MaRDI QIDQ5743136FDOQ5743136
Authors: Robert Tibshirani, Jacob Bien, Noah Simon, Jonathan Taylor, Ryan J. Tibshirani, Jerome H. Friedman, Trevor Hastie 
Publication date: 9 May 2019
Published in: Journal of the Royal Statistical Society Series B: Statistical Methodology (Search for Journal in Brave)
Abstract: We consider rules for discarding predictors in lasso regression and related problems, for computational efficiency. El Ghaoui et al (2010) propose "SAFE" rules that guarantee that a coefficient will be zero in the solution, based on the inner products of each predictor with the outcome. In this paper we propose strong rules that are not foolproof but rarely fail in practice. These can be complemented with simple checks of the Karush- Kuhn-Tucker (KKT) conditions to provide safe rules that offer substantial speed and space savings in a variety of statistical convex optimization problems.
Full work available at URL: https://arxiv.org/abs/1011.2234
Recommendations
- Strong consistency of Lasso estimators
- On LASSO for predictive regression
- On the prediction performance of the Lasso
- The predictive Lasso
- Regularizing LASSO: a consistent variable selection method
- Improving Lasso for model selection and prediction
- A note on the Lasso and related procedures in model selection
- Necessary and sufficient conditions for variable selection consistency of the Lasso in high dimensions
- On the sensitivity of the Lasso to the number of predictor variables
- Lasso penalized model selection criteria for high-dimensional multivariate linear regression analysis
Numerical mathematical programming methods (65K05) Ridge regression; shrinkage estimators (Lasso) (62J07)
Cited In (66)
- An attention algorithm for solving large scale structured \(l_0\)-norm penalty estimation problems
- Fast best subset selection: coordinate descent and local combinatorial optimization algorithms
- Sequential safe feature elimination rule for \(L_1\)-regularized regression with Kullback-Leibler divergence
- A Pliable Lasso
- Title not available (Why is that?)
- Nonsmoothness in machine learning: specific structure, proximal identification, and applications
- Hybrid safe-strong rules for efficient optimization in Lasso-type problems
- Graphical models for zero-inflated single cell gene expression
- Data shared Lasso: a novel tool to discover uplift
- Two-layer feature reduction for sparse-group Lasso via decomposition of convex sets
- On faster convergence of cyclic block coordinate descent-type methods for strongly convex minimization
- Structured variable selection via prior-induced hierarchical penalty functions
- Estimation of semiparametric regression model with right-censored high-dimensional data
- Adaptive hybrid screening for efficient lasso optimization
- Pathwise coordinate optimization for sparse learning: algorithm and theory
- On the distribution, model selection properties and uniqueness of the Lasso estimator in low and high dimensions
- Solving a class of feature selection problems via fractional 0--1 programming
- Lasso screening rules via dual polytope projection
- Gap safe screening rules for sparsity enforcing penalties
- Generalized score matching for non-negative data
- Regularization parameter selection for the low rank matrix recovery
- A dimension reduction technique for large-scale structured sparse optimization problems with application to convex clustering
- Title not available (Why is that?)
- Safe Rules for the Identification of Zeros in the Solutions of the SLOPE Problem
- A coordinate majorization descent algorithm for \(\ell_1\) penalized learning
- A dual based semismooth Newton-type algorithm for solving large-scale sparse Tikhonov regularization problems
- Safe feature screening rules for the regularized Huber regression
- Natural coordinate descent algorithm for \(\ell_1\)-penalised regression in generalised linear models
- Group linear algorithm with sparse principal decomposition: a variable selection and clustering method for generalized linear models
- Best subset, forward stepwise or Lasso? Analysis and recommendations based on extensive comparisons
- Fast stepwise regression based on multidimensional indexes
- Sparse group fused Lasso for model segmentation: a hybrid approach
- \textsf{cmenet}: A new method for bi-level variable selection of conditional main effects
- Title not available (Why is that?)
- Group penalized quantile regression
- Some notes on concordance between optimization and statistics
- Fast and approximate exhaustive variable selection for generalised linear models with APES
- Structured iterative hard thresholding with on- and off-grid applications
- A fast unified algorithm for solving group-lasso penalize learning problems
- Group subset selection for linear regression
- A decomposition method for Lasso problems with zero-sum constraint
- Quantile regression feature selection and estimation with grouped variables using Huber approximation
- PUlasso: High-Dimensional Variable Selection With Presence-Only Data
- Graphical Lasso and thresholding: equivalence and closed-form solutions
- Title not available (Why is that?)
- A safe reinforced feature screening strategy for Lasso based on feasible solutions
- Large-scale multivariate sparse regression with applications to UK Biobank
- Screening rules and its complexity for active set identification
- Scaling up sparse support vector machines by simultaneous feature and sample reduction
- HiQR: an efficient algorithm for high-dimensional quadratic regression with penalties
- Computation and analysis of change points with different jump locations in high-dimensional regression
- ROBOUT: a conditional outlier detection methodology for high-dimensional data
- Techniques for accelerating branch-and-bound algorithms dedicated to sparse optimization
- On stochastic dynamic modeling of incidence data
- Penalized logistic regression with prior information for microarray gene expression classification
- Title not available (Why is that?)
- Locally simultaneous inference
- A safe double screening strategy for elastic net support vector machine
- Algorithms for Sparse Support Vector Machines
- THE LOW-VOLATILITY ANOMALY AND THE ADAPTIVE MULTI-FACTOR MODEL
- A bootstrap model comparison test for identifying genes with context-specific patterns of genetic regulation
- Structure estimation of binary graphical models on stratified data: application to the description of injury tables for victims of road accidents
- A Unified Approach to Sparse Tweedie Modeling of Multisource Insurance Claim Data
- Feature screening strategy for non-convex sparse logistic regression with log sum penalty
- Variable screening for Lasso based on multidimensional indexing
- Lasso regression under stochastic restrictions in linear regression: An application to genomic data
This page was built for publication: Strong Rules for Discarding Predictors in Lasso-Type Problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5743136)
