litlas
ログインアカウント作成

Sparsity and Smoothness Via the Fused Lasso

著者: Robert Tibshirani, Michael A. Saunders, Saharon Rosset, Ji Zhu, Keith Knight - Journal of the Royal Statistical Society Series B (Statistical Methodology) 2004 被引用: 2,805

Summary The lasso penalizes a least squares regression by the sum of the absolute values (L1-norm) of the coefficients. The form of this penalty encourages sparse solutions (with many coefficients equal to 0). We propose the ‘fused lasso’, a generalization that is designed for problems with features that can be ordered in some meaningful way. The fused lasso penalizes the L1-norm of both the coefficients and their successive differences. Thus it encourages sparsity of the coefficients and also sparsity of their differences—i.e. local constancy of the coefficient profile. The fused lasso is especially useful when the number of features p is much greater than N, the sample size. The technique is also extended to the ‘hinge’ loss function that underlies the support vector classifier. We illustrate the methods on examples from protein mass spectroscopy and gene expression data.

Ridge Regression: BiasedEstimation for…Ridge Regression: Biased Estimation for Nonorthogonal ProblemsEstimation of the Meanof a Multivariate Norma…Estimation of the Mean of a Multivariate Normal DistributionA Training Algorithm forOptimal Margin…A Training Algorithm for Optimal Margin ClassifiersIdeal Spatial Adaptationby Wavelet ShrinkageIdeal Spatial Adaptation by Wavelet ShrinkageThe Nature ofStatistical Learning…The Nature of Statistical Learning TheoryRegression Shrinkage andSelection Via the LassoRegression Shrinkage and Selection Via the LassoMolecular Classificationof Cancer: Class…Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression MonitoringAsymptotics forlasso-type estimatorsAsymptotics for lasso-type estimatorsDiagnosis of multiplecancer types by shrunke…Diagnosis of multiple cancer types by shrunken centroids of gene expression1-norm Support VectorMachines1-norm Support Vector MachinesLeast angle regressionLeast angle regressionMulticategory SupportVector MachinesMulticategory Support Vector MachinesPiecewise linearregularized solution…Piecewise linear regularized solution pathsA sparse Bayesianapproach for joint…A sparse Bayesian approach for joint feature selection and classifier learningThe solution path of thegeneralized lassoThe solution path of the generalized lassoOptimization for MachineLearningOptimization for Machine LearningFast Newton-type Methodsfor Total Variation…Fast Newton-type Methods for Total Variation RegularizationAn augmented Lagrangianbased parallel splittin…An augmented Lagrangian based parallel splitting method for separable convex minimization with applications to image processingIdentifying LatentStructures in Panel DataIdentifying Latent Structures in Panel DataAsynchronous DistributedADMM for Large-Scale…Asynchronous Distributed ADMM for Large-Scale Optimization - Part I: Algorithm and Convergence AnalysisOptimal rates for totalvariation denoisingOptimal rates for total variation denoisingDynamic Visualizationand Fast Computation fo…Dynamic Visualization and Fast Computation for Convex Clustering via Algorithmic RegularizationSGL-SVM: A novel methodfor tumor classificatio…SGL-SVM: A novel method for tumor classification via support vector machine with sparse group LassoVector-Valued GraphTrend Filtering With…Vector-Valued Graph Trend Filtering With Non-Convex PenaltiesSparsity and SmoothnessVia the Fused LassoSparsity and Smoothness Via the Fused Lasso過去の参考文献中心の論文この論文を引用する論文古い新しい

ノードをクリックするとフォーカスを固定、空白をクリックすると本論文に戻ります。ホバーで一時的にプレビューできます。各ノードのページはタイトルから開けます。