DASSO: connections between the Dantzig selector and lasso

By Gareth M. James, Peter Radchenko, Jinchi Lv. 2009

196 citations recorded by the source corpus.

Subfields: Computational Mechanics, Statistics and Probability, Signal Processing

View Paper: https://www.semanticscholar.org/paper/ca68550e865cdb68a335d3329d57653da646578d

How researchers see this paper

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

In Computational Mechanics

In Statistics and Probability

In Signal Processing

Citations in the corpus

Regularized rank-based estimation of high-dimensional nonparanormal graphical models (2012, methodology): “However, the resulting estimator shall behave similarly to the rank-based neighborhood Dantzig selector because the lasso penalized least squares and Dantzig selector, in general, work very similarly [Bickel, Ritov and Tsybakov (2009), James, Radchenko and Lv (2009)].”
A Selective Overview of Variable Selection in High Dimensional Feature Space (2009, unknown): “James, Radchenko and Lv (2009) present a formal necessary and suﬃcient condition, as well as easily veriﬁable suﬃcient conditions ensuring the identical solution of the Dantzig selector and the LASSO estimator when the dimensionality p can exceed sample size n .”
Methods for Sparse Signal Recovery Using Kalman Filtering With Embedded Pseudo-Measurement Norms and Quasi-Norms (2010, methodology): “Inspired by the CS approach—while retaining the KF objective function—we replace (4) with the dual problem [11]”
The Discrete Dantzig Selector: Estimating Sparse Linear Models via Mixed Integer Linear Optimization (2015, background): “sparse linear models, as the data-fidelity parameter is allowed to vary [8], and allows a natural extension to more general response distributions [13].”
The solution path of the generalized lasso (2010, unknown): “The ﬁrst of its kind, LARS inspired the development of path algorithms for various other optimization problems that appear in statistics [16, 25, 19, 20], and our case is no exception.”
Feature Selection for Classification: A Review (2014, background): “Strong theoretical results show that LASSO and Dantzig selector are closely related [28].”
A generalized Dantzig selector with shrinkage tuning (2009, methodology): “…the above procedure is simplified by removing Steps 2 and 3, and using the matrix B that is produced by selecting all the columns of A that correspond to the non-zero components of β + and β − ; see James et al. (2008a) for further explanation of, and motivation for, the direction step.”
Some sharp performance bounds for least squares regression with L1 regularization (2009, result): “However, it was observed [10, 8, 13, 12] that the performance of Lasso is comparable to that of the Dantzig selector.”
Feature Selection (2016, background): “In (Candes and Tao, 2007; James et al., 2009), the authors show that the errors of Dantzig selector is up to a logarithmic factor log(d) (d is the feature dimensionality).”
Templates for convex cone problems with applications to sparse signal recovery (2010, methodology): “2) is via the new works [38, 58], which use homotopy methods inspired by LARS to compute the whole solution path of the Dantzig selector.”
Generalized Dantzig Selector: Application to the k-support norm (2014, methodology): “For the L1-norm Dantzig selector, [3] proposed a primal-dual interior point method since the optimization is a linear program.”