We consider the problem of estimating a function f0 in logistic regression model. We propose to estimate this function f0 by a sparse approximation build as a linear combination of elements of a given dictionary of p functions. This sparse approximation is selected by the Lasso or Group Lasso procedure. In this context, we state non asymptotic oracle inequalities for Lasso and Group Lasso under...

Abstract: We revisit the adaptive Lasso as well as the thresholded Lasso with refitting, in a high-dimensional linear model, and study prediction error, lq-error (q ∈ {1, 2}), and number of false positive selections. Our theoretical results for the two methods are, at a rather fine scale, comparable. The differences only show up in terms of the (minimal) restricted and sparse eigenvalues, favor...

Given n noisy samples with p dimensions, where n " p, we show that the multi-step thresholding procedure based on the Lasso – we call it the Thresholded Lasso, can accurately estimate a sparse vector β ∈ R in a linear model Y = Xβ + ", where Xn×p is a design matrix normalized to have column #2-norm √ n, and " ∼ N(0,σIn). We show that under the restricted eigenvalue (RE) condition (BickelRitov-T...

This note extends an attribute of the LASSO procedure to a whole class related procedures, including square-root LASSO, square LAD-LASSO, and instance generalized LASSO. Namely, under assumption that input matrix satisfies ℓ p -restricted isometry property (which in some sense is weaker than standard 2 assumption), it shown if vector comes from exact measurement sparse vector, then minimizer an...

Partial linear model is very flexible when the relation between the covariates and responses, either parametric and nonparametric. However, estimation of the regression coefficients is challenging since one must also estimate the nonparametric component simultaneously. As a remedy, the differencing approach, to eliminate the nonparametric component and estimate the regression coefficients, can ...

This paper introduce simple and general theories of compressed sensing and LASSO. The novelty is that our recovery results do not require the restricted isometry property(RIP). We use the notion of weak RIP that is a natural generalization of RIP. We consider that the proposed results are more useful and flexible for real data analysis in various fields.

This paper shows that the least absolute shrinkage and selection operator (LASSO) can provide an exact optimal solution to a special type of constrained cardinality minimization problem, which is motivated from a sensor network measurement robustness analysis problem. The constraint matrix of the considered problem is totally unimodular. This is shown to imply that LASSO leads to a tight linear...

Oracle inequalities and variable selection properties for the Lasso in linear models have been established under a variety of different assumptions on the design matrix. We show in this paper how the different conditions and concepts relate to each other. The restricted eigenvalue condition (Bickel et al., 2009) or the slightly weaker compatibility condition (van de Geer, 2007) are sufficient f...

In the high-dimensional regression model a response variable is linearly related to p covariates, but the sample size n is smaller than p. We assume that only a small subset of covariates is ‘active’ (i.e., the corresponding coefficients are non-zero), and consider the model-selection problem of identifying the active covariates. A popular approach is to estimate the regression coefficients thr...