We study feature selection in high-dimensional regression under two distinct sources of instability: sampling variability and measurement error in the design matrix. Stability Selection addresses the former through sub-sampling and aggregation, but does not explicitly stress-test robustness to noisy predictors. We introduce doubly stable feature selection, a perturb-and-aggregate framework that targets features whose inclusion is stable both across randomization and across increasing levels of design noise. The method injects controlled additive noise into the design matrix, fits a fixed base selector such as the Lasso on the perturbed data, and aggregates selection frequencies. Sweeping over a grid of noise levels yields a stability path that summarizes robustness to measurement error while using the full sample size and isolating the effect of design perturbations. On the theory side, we show that classical model-selection conditions are preserved under sufficiently small perturbations, with a high-probability extension for Gaussian noise. Empirically, experiments on synthetic and real datasets show improved robustness compared with Stability Selection and standard base selectors.

In this section, we describe Nadaraya-Watson (NW) classifier, Local Polynomial (LP) classifier and their convergence rates (Audibert & Tsybakov, 2007). In what follows,K: X R represents a kernel function, e.g., Gaussian kernel K(X):=exp( kXk22),andh>0representsabandwidth. LP classifier is thus proved to be an optimal classifier in this sense. The two error terms are in fact combined asδβ,r(X) = O(rβ), because 2bβ/2c+2 β. In step (i), queries are first classified into two different cases, i.e., (X) io and (X) > io.

Score estimation has recently emerged as a key modern statistical challenge, due to its pivotal role in generative modelling via diffusion models. Moreover, it is an essential ingredient in a new approach to linear regression via convex $M$-estimation, where the corresponding error densities are projected onto the log-concave class. Motivated by these applications, we study the minimax risk of score estimation with respect to squared $L^2(P_0)$-loss, where $P_0$ denotes an underlying log-concave distribution on $\mathbb{R}$. Such distributions have decreasing score functions, but on its own, this shape constraint is insufficient to guarantee a finite minimax risk. We therefore define subclasses of log-concave densities that capture two fundamental aspects of the estimation problem. First, we establish the crucial impact of tail behaviour on score estimation by determining the minimax rate over a class of log-concave densities whose score function exhibits controlled growth relative to the quantile levels. Second, we explore the interplay between smoothness and log-concavity by considering the class of log-concave densities with a scale restriction and a $(β,L)$-Hölder assumption on the log-density for some $β\in [1,2]$. We show that the minimax risk over this latter class is of order $L^{2/(2β+1)}n^{-β/(2β+1)}$ up to poly-logarithmic factors, where $n$ denotes the sample size. When $β< 2$, this rate is faster than could be obtained under either the shape constraint or the smoothness assumption alone. Our upper bounds are attained by a locally adaptive, multiscale estimator constructed from a uniform confidence band for the score function. This study highlights intriguing differences between the score estimation and density estimation problems over this shape-constrained class.

We survey the field of nonparametric inference under shape constraints, providing a historical overview and a perspective on its current state. An outlook and some open problems offer thoughts on future directions. 1 Introduction. Traditionally, we think of statistical methods as being divided into parametric approaches, which can be restrictive, but where estimation is typically straightforward (e.g. using maximum likelihood), and nonparametric methods, which are more flexible but often require careful choices of tuning parameters. Nonparametric inference under shape constraints sits somewhere in the middle, seeking in some ways the best of both worlds. The origins of the field are often traced to Grenander (1956), who proved that there exists a unique maximum likelihood estimator (MLE) of a decreasing density on the non-negative half-line (and was able to characterise it explicitly).

In this section, we describe Nadaraya-Watson (NW) classifier, Local Polynomial (LP) classifier and their convergence rates (Audibert & Tsybakov, 2007). Proof of Corollary 2. Proposition 6 immediately proves the assertion. We basically follow the proof of Chaudhuri & Dasgupta (2014) Theorem 4(b). In Section G.1, we first define symbols In Section G.2, we describe the sketch of the proof and main differences between our proof and that of Section G.3 shows the main body of the Proof, by utilizing several Lemmas listed in A minimum radius whose measure of the ball is larger than t > 0, i.e., r Chaudhuri & Dasgupta (2014) Lemma 21) Then, the assertion is proved. See the following Section G.4 for Lemma 1-7 used in this proof.

R2: It would benefit the paper greatly to provide some more discussion with the conditions in Theorem 2.

Shape constraints yield flexible middle grounds between fully nonparametric and fully parametric approaches to modeling distributions of data. The specific assumption of log-concavity is motivated by applications across economics, survival modeling, and reliability theory. However, there do not currently exist valid tests for whether the underlying density of given data is log-concave. The recent universal inference methodology provides a valid test. The universal test relies on maximum likelihood estimation (MLE), and efficient methods already exist for finding the log-concave MLE. This yields the first test of log-concavity that is provably valid in finite samples in any dimension, for which we also establish asymptotic consistency results. Empirically, we find that the highest power is obtained by using random projections to convert the d-dimensional testing problem into many one-dimensional problems, leading to a simple procedure that is statistically and computationally efficient.