Searcharxiv⌕ Search

arXiv subjects

Adityanand Guntuboyina

Publications and source records attributed to Adityanand Guntuboyina.

36 records · Page 2Linked to original sources
arXiv2018

Adaptive Risk Bounds in Univariate Total Variation Denoising and Trend Filtering

We study trend filtering, a relatively recent method for univariate nonparametric regression. For a given positive integer $r$, the $r$-th order trend filtering estimator is defined as the minimizer of the sum of squared errors when we constrain (or penalize) the sum of the absolute $r$-th order discrete derivatives of the fitted function at the design points. For $r=1$, the estimator reduces to total variation regularization which has received much attention in the statistics and image processing literature. In this paper, we study the performance of the trend filtering estimator for every positive integer $r$, both in the constrained and penalized forms. Our main results show that in the strong sparsity setting when the underlying function is a (discrete) spline with few "knots", the risk (under the global squared error loss) of the trend filtering estimator (with an appropriate choice of the tuning parameter) achieves the parametric $n^{-1}$ rate, up to a logarithmic (multiplicative) factor. Our results therefore provide support for the use of trend filtering, for every $r$, in the strong sparsity setting.

math.ST↗
arXiv2018

The geometry of hypothesis testing over convex cones: Generalized likelihood tests and minimax radii

We consider a compound testing problem within the Gaussian sequence model in which the null and alternative are specified by a pair of closed, convex cones. Such cone testing problem arise in various applications, including detection of treatment effects, trend detection in econometrics, signal detection in radar processing, and shape-constrained inference in non-parametric statistics. We provide a sharp characterization of the GLRT testing radius up to a universal multiplicative constant in terms of the geometric structure of the underlying convex cones. When applied to concrete examples, this result reveals some interesting phenomena that do not arise in the analogous problems of estimation under convex constraints. In particular, in contrast to estimation error, the testing error no longer depends purely on the problem complexity via a volume-based measure (such as metric entropy or Gaussian complexity), other geometric properties of the cones also play an important role. To address the issue of optimality, we prove information-theoretic lower bounds for minimax testing radius again in terms of geometric quantities. Our general theorems are illustrated by examples including the cases of monotone and orthant cones, and involve some results of independent interest.

math.ST↗
arXiv2017

On the risk of convex-constrained least squares estimators under misspecification

We consider the problem of estimating the mean of a noisy vector. When the mean lies in a convex constraint set, the least squares projection of the random vector onto the set is a natural estimator. Properties of the risk of this estimator, such as its asymptotic behavior as the noise tends to zero, have been well studied. We instead study the behavior of this estimator under misspecification, that is, without the assumption that the mean lies in the constraint set. For appropriately defined notions of risk in the misspecified setting, we prove a generalization of a low noise characterization of the risk due to Oymak and Hassibi in the case of a polyhedral constraint set. An interesting consequence of our results is that the risk can be much smaller in the misspecified setting than in the well-specified setting. We also discuss consequences of our result for isotonic regression.

math.ST↗
arXiv2017

A note on the approximate admissibility of regularized estimators in the Gaussian sequence model

We study the problem of estimating an unknown vector $θ$ from an observation $X$ drawn according to the normal distribution with mean $θ$ and identity covariance matrix under the knowledge that $θ$ belongs to a known closed convex set $Θ$. In this general setting, Chatterjee (2014) proved that the natural constrained least squares estimator is "approximately admissible" for every $Θ$. We extend this result by proving that the same property holds for all convex penalized estimators as well. Moreover, we simplify and shorten the original proof considerably. We also provide explicit upper and lower bounds for the universal constant underlying the notion of approximate admissibility.

math.ST↗
arXiv2016

On Bayes Risk Lower Bounds

This paper provides a general technique for lower bounding the Bayes risk of statistical estimation, applicable to arbitrary loss functions and arbitrary prior distributions. A lower bound on the Bayes risk not only serves as a lower bound on the minimax risk, but also characterizes the fundamental limit of any estimator given the prior knowledge. Our bounds are based on the notion of $f$-informativity, which is a function of the underlying class of probability measures and the prior. Application of our bounds requires upper bounds on the $f$-informativity, thus we derive new upper bounds on $f$-informativity which often lead to tight Bayes risk lower bounds. Our technique leads to generalizations of a variety of classical minimax bounds (e.g., generalized Fano's inequality). Our Bayes risk lower bounds can be directly applied to several concrete estimation problems, including Gaussian location models, generalized linear models, and principal component analysis for spiked covariance models. To further demonstrate the applications of our Bayes risk lower bounds to machine learning problems, we present two new theoretical results: (1) a precise characterization of the minimax risk of learning spherical Gaussian mixture models under the smoothed analysis framework, and (2) lower bounds for the Bayes risk under a natural prior for both the prediction and estimation errors for high-dimensional sparse linear regression under an improper learning setting.

math.ST↗
arXiv2016

Stochastically Transitive Models for Pairwise Comparisons: Statistical and Computational Issues

There are various parametric models for analyzing pairwise comparison data, including the Bradley-Terry-Luce (BTL) and Thurstone models, but their reliance on strong parametric assumptions is limiting. In this work, we study a flexible model for pairwise comparisons, under which the probabilities of outcomes are required only to satisfy a natural form of stochastic transitivity. This class includes parametric models including the BTL and Thurstone models as special cases, but is considerably more general. We provide various examples of models in this broader stochastically transitive class for which classical parametric models provide poor fits. Despite this greater flexibility, we show that the matrix of probabilities can be estimated at the same rate as in standard parametric models. On the other hand, unlike in the BTL and Thurstone models, computing the minimax-optimal estimator in the stochastically transitive model is non-trivial, and we explore various computationally tractable alternatives. We show that a simple singular value thresholding algorithm is statistically consistent but does not achieve the minimax rate. We then propose and study algorithms that achieve the minimax rate over interesting sub-classes of the full stochastically transitive class. We complement our theoretical results with thorough numerical simulations.

stat.ML↗
arXiv2016

Adaptation in log-concave density estimation

The log-concave maximum likelihood estimator of a density on the real line based on a sample of size $n$ is known to attain the minimax optimal rate of convergence of $O(n^{-4/5})$ with respect to, e.g., squared Hellinger distance. In this paper, we show that it also enjoys attractive adaptation properties, in the sense that it achieves a faster rate of convergence when the logarithm of the true density is $k$-affine (i.e.\ made up of $k$ affine pieces), provided $k$ is not too large. Our results use two different techniques: the first relies on a new Marshall's inequality for log-concave density estimation, and reveals that when the true density is close to log-linear on its support, the log-concave maximum likelihood estimator can achieve the parametric rate of convergence in total variation distance. Our second approach depends on local bracketing entropy methods, and allows us to prove a sharp oracle inequality, which implies in particular that the rate of convergence with respect to various global loss functions, including Kullback--Leibler divergence, is $O\bigl(\frac{k}{n}\log^{5/4} n\bigr)$ when the true density is log-concave and its logarithm is close to $k$-affine.

math.ST↗
arXiv2015

On matrix estimation under monotonicity constraints

We consider the problem of estimating an unknown $n_1 \times n_2$ matrix $\mathbf{θ^*}$ from noisy observations under the constraint that $\mathbfθ^*$ is nondecreasing in both rows and columns. We consider the least squares estimator (LSE) in this setting and study its risk properties. We show that the worst case risk of the LSE is $n^{-1/2}$, up to multiplicative logarithmic factors, where $n = n_1 n_2$ and that the LSE is minimax rate optimal (up to logarithmic factors). We further prove that for some special $\mathbfθ^*$, the risk of the LSE could be much smaller than $n^{-1/2}$; in fact, it could even be parametric i.e., $n^{-1}$ up to logarithmic factors. Such parametric rates occur when the number of "rectangular" blocks of $\mathbfθ^*$ is bounded from above by a constant. We derive, as a consequence, an interesting adaptation property of the LSE which we term variable adaptation -- the LSE performs as well as the oracle estimator when estimating a matrix that is constant along each row/column. Our proofs borrow ideas from empirical process theory and convex geometry and are of independent interest.

math.ST↗
arXiv2015

Adaptive estimation of planar convex sets

In this paper, we consider adaptive estimation of an unknown planar compact, convex set from noisy measurements of its support function on a uniform grid. Both the problem of estimating the support function at a point and that of estimating the convex set are studied. Data-driven adaptive estimators are proposed and their optimality properties are established. For pointwise estimation, it is shown that the estimator optimally adapts to every compact, convex set instead of a collection of large parameter spaces as in the conventional minimax theory of nonparametric estimation. For set estimation, the estimators adaptively achieve the optimal rate of convergence. In both these problems, our analysis makes no smoothness assumptions on the unknown sets.

math.ST↗
arXiv2015

On risk bounds in isotonic and other shape restricted regression problems

We consider the problem of estimating an unknown $θ\in {\mathbb{R}}^n$ from noisy observations under the constraint that $θ$ belongs to certain convex polyhedral cones in ${\mathbb{R}}^n$. Under this setting, we prove bounds for the risk of the least squares estimator (LSE). The obtained risk bound behaves differently depending on the true sequence $θ$ which highlights the adaptive behavior of $θ$. As special cases of our general result, we derive risk bounds for the LSE in univariate isotonic and convex regression. We study the risk bound in isotonic regression in greater detail: we show that the isotonic LSE converges at a whole range of rates from $\log n/n$ (when $θ$ is constant) to $n^{-2/3}$ (when $θ$ is uniformly increasing in a certain sense). We argue that the bound presents a benchmark for the risk of any estimator in isotonic regression by proving nonasymptotic local minimax lower bounds. We prove an analogue of our bound for model misspecification where the true $θ$ is not necessarily nondecreasing.

math.ST↗
arXiv2014

Global risk bounds and adaptation in univariate convex regression

We consider the problem of nonparametric estimation of a convex regression function $ϕ_0$. We study the risk of the least squares estimator (LSE) under the natural squared error loss. We show that the risk is always bounded from above by $n^{-4/5}$ modulo logarithmic factors while being much smaller when $ϕ_0$ is well-approximable by a piecewise affine convex function with not too many affine pieces (in which case, the risk is at most $1/n$ up to logarithmic factors). On the other hand, when $ϕ_0$ has curvature, we show that no estimator can have risk smaller than a constant multiple of $n^{-4/5}$ in a very strong sense by proving a "local" minimax lower bound. We also study the case of model misspecification where we show that the LSE exhibits the same global behavior provided the loss is measured from the closest convex projection of the true regression function. In the process of deriving our risk bounds, we prove new results for the metric entropy of local neighborhoods of the space of univariate convex functions. These results, which may be of independent interest, demonstrate the non-uniform nature of the space of univariate convex functions in sharp contrast to classical function spaces based on smoothness constraints.

math.ST↗
arXiv2014

On the impossibility of constructing good population mean estimators in a realistic Respondent Driven Sampling model

Current methods for population mean estimation from data collected by Respondent Driven Sampling (RDS) are based on the Horvitz-Thompson estimator together with a set of assumptions on the sampling model under which the inclusion probabilities can be determined from the information contained in the data. In this paper, we argue that such set of assumptions are too simplistic to be realistic and that under realistic sampling models, the situation is far more complicated. Specifically, we study a realistic RDS sampling model that is motivated by a real world RDS dataset. We show that, for this model, the inclusion probabilities, which are necessary for the application of the Horvitz-Thompson estimator, can not be determined by the information in the sample alone. An implication is that, unless additional information about the underlying population network is obtained, it is hopeless to conceive of a general theory of population mean estimation from current RDS data.

stat.ME↗
arXiv2014

Covering numbers of $L_p$-balls of convex sets and functions

We prove bounds for the covering numbers of classes of convex functions and convex sets in Euclidean space. Previous results require the underlying convex functions or sets to be uniformly bounded. We relax this assumption and replace it with weaker integral constraints. Existing results can be recovered as special cases of our results.

cs.IT↗
arXiv2013

Sharp Inequalities for $f$-divergences

$f$-divergences are a general class of divergences between probability measures which include as special cases many commonly used divergences in probability, mathematical statistics and information theory such as Kullback-Leibler divergence, chi-squared divergence, squared Hellinger distance, total variation distance etc. In this paper, we study the problem of maximizing or minimizing an $f$-divergence between two probability measures subject to a finite number of constraints on other $f$-divergences. We show that these infinite-dimensional optimization problems can all be reduced to optimization problems over small finite dimensional spaces which are tractable. Our results lead to a comprehensive and unified treatment of the problem of obtaining sharp inequalities between $f$-divergences. We demonstrate that many of the existing results on inequalities between $f$-divergences can be obtained as special cases of our results and we also improve on some existing non-sharp inequalities.

math.ST↗
arXiv2012

Covering Numbers for Convex Functions

In this paper we study the covering numbers of the space of convex and uniformly bounded functions in multi-dimension. We find optimal upper and lower bounds for the $ε$-covering number of $\C([a, b]^d, B)$, in the $L_p$-metric, $1 \le p < \infty$, in terms of the relevant constants, where $d \geq 1$, $a < b \in \mathbb{R}$, $B>0$, and $\C([a,b]^d, B)$ denotes the set of all convex functions on $[a, b]^d$ that are uniformly bounded by $B$. We summarize previously known results on covering numbers for convex functions and also provide alternate proofs of some known results. Our results have direct implications in the study of rates of convergence of empirical minimization procedures as well as optimal convergence rates in the numerous convexity constrained function estimation problems.

cs.IT↗
arXiv2011

Lower bounds for the minimax risk using $f$-divergences and applications

Lower bounds involving $f$-divergences between the underlying probability measures are proved for the minimax risk in estimation problems. Our proofs just use simple convexity facts. Special cases and straightforward corollaries of our bounds include well known inequalities for establishing minimax lower bounds such as Fano's inequality, Pinsker's inequality and inequalities based on global entropy conditions. Two applications are provided: a new minimax lower bound for the reconstruction of convex bodies from noisy support function measurements and a different proof of a recent minimax lower bound for the estimation of a covariance matrix.

math.ST↗