arXiv subjects

Martin J. Wainwright

Publications and source records attributed to Martin J. Wainwright.

At least 91 records · Page 5Linked to original sources

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↗

arXiv2018

From Gauss to Kolmogorov: Localized Measures of Complexity for Ellipses

The Gaussian width is a fundamental quantity in probability, statistics and geometry, known to underlie the intrinsic difficulty of estimation and hypothesis testing. In this work, we show how the Gaussian width, when localized to any given point of an ellipse, can be controlled by the Kolmogorov width of a set similarly localized. This connection leads to an explicit characterization of the estimation error of least-squares regression as a function of the true regression vector within the ellipse. The rate of error decay varies substantially as a function of location: as a concrete example, in Sobolev ellipses of smoothness $α$, we exhibit rates that vary from $(σ^2)^{\frac{2 α}{2 α+ 1}}$, corresponding to the classical global rate, to the faster rate $(σ^2)^{\frac{4 α}{4 α+ 1}}$. We also show how the local Kolmogorov width can be related to local metric entropy.

math.ST↗

arXiv2018

Early stopping for kernel boosting algorithms: A general analysis with localized complexities

Early stopping of iterative algorithms is a widely-used form of regularization in statistics, commonly used in conjunction with boosting and related gradient-type algorithms. Although consistency results have been established in some settings, such estimators are less well-understood than their analogues based on penalized regularization. In this paper, for a relatively broad class of loss functions and boosting algorithms (including L2-boost, LogitBoost and AdaBoost, among others), we exhibit a direct connection between the performance of a stopped iterate and the localized Gaussian complexity of the associated function class. This connection allows us to show that local fixed point analysis of Gaussian or Rademacher complexities, now standard in the analysis of penalized estimators, can be used to derive optimal stopping rules. We derive such stopping rules in detail for various kernel classes, and illustrate the correspondence of our theory with practice for Sobolev kernel classes.

stat.ML↗

arXiv2018

Approximate Ranking from Pairwise Comparisons

A common problem in machine learning is to rank a set of n items based on pairwise comparisons. Here ranking refers to partitioning the items into sets of pre-specified sizes according to their scores, which includes identification of the top-k items as the most prominent special case. The score of a given item is defined as the probability that it beats a randomly chosen other item. Finding an exact ranking typically requires a prohibitively large number of comparisons, but in practice, approximate rankings are often adequate. Accordingly, we study the problem of finding approximate rankings from pairwise comparisons. We analyze an active ranking algorithm that counts the number of comparisons won, and decides whether to stop or which pair of items to compare next, based on confidence intervals computed from the data collected in previous steps. We show that this algorithm succeeds in recovering approximate rankings using a number of comparisons that is close to optimal up to logarithmic factors. We also present numerical results, showing that in practice, approximation can drastically reduce the number of comparisons required to estimate a ranking.

cs.LG↗

arXiv2018

The local geometry of testing in ellipses: Tight control via localized Kolmogorov widths

We study the local geometry of testing a mean vector within a high-dimensional ellipse against a compound alternative. Given samples of a Gaussian random vector, the goal is to distinguish whether the mean is equal to a known vector within an ellipse, or equal to some other unknown vector in the ellipse. Such ellipse testing problems lie at the heart of several applications, including non-parametric goodness-of-fit testing, signal detection in cognitive radio, and regression function testing in reproducing kernel Hilbert spaces. While past work on such problems has focused on the difficulty in a global sense, we study difficulty in a way that is localized to each vector within the ellipse. Our main result is to give sharp upper and lower bounds on the localized minimax testing radius in terms of an explicit formula involving the Kolmogorov width of the ellipse intersected with a Euclidean ball. When applied to particular examples, our general theorems yield interesting rates that were not known before: as a particular case, for testing in Sobolev ellipses of smoothness $α$, we demonstrate rates that vary from $(σ^2)^{\frac{4 α}{4 α+ 1}}$, corresponding to the classical global rate, to the faster rate $(σ^2)^{\frac{8 α}{8 α+ 1}}$, achievable for vectors at favorable locations within the ellipse. We also show that the optimal test for this problem is achieved by a linear projection test that is based on an explicit lower-dimensional projection of the observation vector.

math.ST↗

arXiv2017

A framework for Multi-A(rmed)/B(andit) testing with online FDR control

We propose an alternative framework to existing setups for controlling false alarms when multiple A/B tests are run over time. This setup arises in many practical applications, e.g. when pharmaceutical companies test new treatment options against control pills for different diseases, or when internet companies test their default webpages versus various alternatives over time. Our framework proposes to replace a sequence of A/B tests by a sequence of best-arm MAB instances, which can be continuously monitored by the data scientist. When interleaving the MAB tests with an an online false discovery rate (FDR) algorithm, we can obtain the best of both worlds: low sample complexity and any time online FDR control. Our main contributions are: (i) to propose reasonable definitions of a null hypothesis for MAB instances; (ii) to demonstrate how one can derive an always-valid sequential p-value that allows continuous monitoring of each MAB test; and (iii) to show that using rejection thresholds of online-FDR algorithms as the confidence levels for the MAB algorithms results in both sample-optimality, high power and low FDR at any point in time. We run extensive simulations to verify our claims, and also report results on real data collected from the New Yorker Cartoon Caption contest.

stat.ML↗

arXiv2017

Online control of the false discovery rate with decaying memory

In the online multiple testing problem, p-values corresponding to different null hypotheses are observed one by one, and the decision of whether or not to reject the current hypothesis must be made immediately, after which the next p-value is observed. Alpha-investing algorithms to control the false discovery rate (FDR), formulated by Foster and Stine, have been generalized and applied to many settings, including quality-preserving databases in science and multiple A/B or multi-armed bandit tests for internet commerce. This paper improves the class of generalized alpha-investing algorithms (GAI) in four ways: (a) we show how to uniformly improve the power of the entire class of monotone GAI procedures by awarding more alpha-wealth for each rejection, giving a win-win resolution to a recent dilemma raised by Javanmard and Montanari, (b) we demonstrate how to incorporate prior weights to indicate domain knowledge of which hypotheses are likely to be non-null, (c) we allow for differing penalties for false discoveries to indicate that some hypotheses may be more important than others, (d) we define a new quantity called the decaying memory false discovery rate (mem-FDR) that may be more meaningful for truly temporal applications, and which alleviates problems that we describe and refer to as "piggybacking" and "alpha-death". Our GAI++ algorithms incorporate all four generalizations simultaneously, and reduce to more powerful variants of earlier algorithms when the weights and decay are all set to unity. Finally, we also describe a simple method to derive new online FDR rules based on an estimated false discovery proportion.

stat.ME↗

arXiv2017

Low Permutation-rank Matrices: Structural Properties and Noisy Completion

We consider the problem of noisy matrix completion, in which the goal is to reconstruct a structured matrix whose entries are partially observed in noise. Standard approaches to this underdetermined inverse problem are based on assuming that the underlying matrix has low rank, or is well-approximated by a low rank matrix. In this paper, we propose a richer model based on what we term the "permutation-rank" of a matrix. We first describe how the classical non-negative rank model enforces restrictions that may be undesirable in practice, and how and these restrictions can be avoided by using the richer permutation-rank model. Second, we establish the minimax rates of estimation under the new permutation-based model, and prove that surprisingly, the minimax rates are equivalent up to logarithmic factors to those for estimation under the typical low rank model. Third, we analyze a computationally efficient singular-value-thresholding algorithm, known to be optimal for the low-rank setting, and show that it also simultaneously yields a consistent estimator for the low-permutation rank setting. Finally, we present various structural results characterizing the uniqueness of the permutation-rank decomposition, and characterizing convex approximations of the permutation-rank polytope.

stat.ML↗

arXiv2017

On kernel methods for covariates that are rankings

Permutation-valued features arise in a variety of applications, either in a direct way when preferences are elicited over a collection of items, or an indirect way in which numerical ratings are converted to a ranking. To date, there has been relatively limited study of regression, classification, and testing problems based on permutation-valued features, as opposed to permutation-valued responses. This paper studies the use of reproducing kernel Hilbert space methods for learning from permutation-valued features. These methods embed the rankings into an implicitly defined function space, and allow for efficient estimation of regression and test functions in this richer space. Our first contribution is to characterize both the feature spaces and spectral properties associated with two kernels for rankings, the Kendall and Mallows kernels. Using tools from representation theory, we explain the limited expressive power of the Kendall kernel by characterizing its degenerate spectrum, and in sharp contrast, we prove that Mallows' kernel is universal and characteristic. We also introduce families of polynomial kernels that interpolate between the Kendall (degree one) and Mallows' (infinite degree) kernels. We show the practical effectiveness of our methods via applications to Eurobarometer survey data as well as a Movielens ratings dataset.

stat.ML↗

arXiv2017

Worst-case vs Average-case Design for Estimation from Fixed Pairwise Comparisons

Pairwise comparison data arises in many domains, including tournament rankings, web search, and preference elicitation. Given noisy comparisons of a fixed subset of pairs of items, we study the problem of estimating the underlying comparison probabilities under the assumption of strong stochastic transitivity (SST). We also consider the noisy sorting subclass of the SST model. We show that when the assignment of items to the topology is arbitrary, these permutation-based models, unlike their parametric counterparts, do not admit consistent estimation for most comparison topologies used in practice. We then demonstrate that consistent estimation is possible when the assignment of items to the topology is randomized, thus establishing a dichotomy between worst-case and average-case designs. We propose two estimators in the average-case setting and analyze their risk, showing that it depends on the comparison topology only through the degree sequence of the topology. The rates achieved by these estimators are shown to be optimal for a large class of graphs. Our results are corroborated by simulations on multiple comparison topologies.

cs.LG↗

arXiv2017

Optimal Rates and Tradeoffs in Multiple Testing

Multiple hypothesis testing is a central topic in statistics, but despite abundant work on the false discovery rate (FDR) and the corresponding Type-II error concept known as the false non-discovery rate (FNR), a fine-grained understanding of the fundamental limits of multiple testing has not been developed. Our main contribution is to derive a precise non-asymptotic tradeoff between FNR and FDR for a variant of the generalized Gaussian sequence model. Our analysis is flexible enough to permit analyses of settings where the problem parameters vary with the number of hypotheses $n$, including various sparse and dense regimes (with $o(n)$ and $\mathcal{O}(n)$ signals). Moreover, we prove that the Benjamini-Hochberg algorithm as well as the Barber-Candès algorithm are both rate-optimal up to constants across these regimes.

math.ST↗

arXiv2017

Denoising Linear Models with Permuted Data

The multivariate linear regression model with shuffled data and additive Gaussian noise arises in various correspondence estimation and matching problems. Focusing on the denoising aspect of this problem, we provide a characterization the minimax error rate that is sharp up to logarithmic factors. We also analyze the performance of two versions of a computationally efficient estimator, and establish their consistency for a large range of input parameters. Finally, we provide an exact algorithm for the noiseless problem and demonstrate its performance on an image point-cloud matching task. Our analysis also extends to datasets with outliers.

stat.ML↗

arXiv2016

Function-Specific Mixing Times and Concentration Away from Equilibrium

Slow mixing is the central hurdle when working with Markov chains, especially those used for Monte Carlo approximations (MCMC). In many applications, it is only of interest to estimate the stationary expectations of a small set of functions, and so the usual definition of mixing based on total variation convergence may be too conservative. Accordingly, we introduce function-specific analogs of mixing times and spectral gaps, and use them to prove Hoeffding-like function-specific concentration inequalities. These results show that it is possible for empirical expectations of functions to concentrate long before the underlying chain has mixed in the classical sense, and we show that the concentration rates we achieve are optimal up to constants. We use our techniques to derive confidence intervals that are sharper than those implied by both classical Markov chain Hoeffding bounds and Berry-Esseen-corrected CLT bounds. For applications that require testing, rather than point estimation, we show similar improvements over recent sequential testing results for MCMC. We conclude by applying our framework to real data examples of MCMC, providing evidence that our theory is both accurate and relevant to practice.

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

Active Ranking from Pairwise Comparisons and when Parametric Assumptions Don't Help

We consider sequential or active ranking of a set of n items based on noisy pairwise comparisons. Items are ranked according to the probability that a given item beats a randomly chosen item, and ranking refers to partitioning the items into sets of pre-specified sizes according to their scores. This notion of ranking includes as special cases the identification of the top-k items and the total ordering of the items. We first analyze a sequential ranking algorithm that counts the number of comparisons won, and uses these counts to decide whether to stop, or to compare another pair of items, chosen based on confidence intervals specified by the data collected up to that point. We prove that this algorithm succeeds in recovering the ranking using a number of comparisons that is optimal up to logarithmic factors. This guarantee does not require any structural properties of the underlying pairwise probability matrix, unlike a significant body of past work on pairwise ranking based on parametric models such as the Thurstone or Bradley-Terry-Luce models. It has been a long-standing open question as to whether or not imposing these parametric assumptions allows for improved ranking algorithms. For stochastic comparison models, in which the pairwise probabilities are bounded away from zero, our second contribution is to resolve this issue by proving a lower bound for parametric models. This shows, perhaps surprisingly, that these popular parametric modeling choices offer at most logarithmic gains for stochastic comparisons.

cs.LG↗

arXiv2016

Local Maxima in the Likelihood of Gaussian Mixture Models: Structural Results and Algorithmic Consequences

We provide two fundamental results on the population (infinite-sample) likelihood function of Gaussian mixture models with $M \geq 3$ components. Our first main result shows that the population likelihood function has bad local maxima even in the special case of equally-weighted mixtures of well-separated and spherical Gaussians. We prove that the log-likelihood value of these bad local maxima can be arbitrarily worse than that of any global optimum, thereby resolving an open question of Srebro (2007). Our second main result shows that the EM algorithm (or a first-order variant of it) with random initialization will converge to bad critical points with probability at least $1-e^{-Ω(M)}$. We further establish that a first-order variant of EM will not converge to strict saddle points almost surely, indicating that the poor performance of the first-order method can be attributed to the existence of bad local maxima rather than bad saddle points. Overall, our results highlight the necessity of careful initialization when using the EM algorithm in practice, even when applied in highly favorable settings.

stat.ML↗

arXiv2016

Convexified Convolutional Neural Networks

We describe the class of convexified convolutional neural networks (CCNNs), which capture the parameter sharing of convolutional neural networks in a convex manner. By representing the nonlinear convolutional filters as vectors in a reproducing kernel Hilbert space, the CNN parameters can be represented as a low-rank matrix, which can be relaxed to obtain a convex optimization problem. For learning two-layer convolutional neural networks, we prove that the generalization error obtained by a convexified CNN converges to that of the best possible CNN. For learning deeper networks, we train CCNNs in a layer-wise manner. Empirically, CCNNs achieve performance competitive with CNNs trained by backpropagation, SVMs, fully-connected neural networks, stacked denoising auto-encoders, and other baseline methods.

cs.LG↗

arXiv2016

Linear Regression with an Unknown Permutation: Statistical and Computational Limits

Consider a noisy linear observation model with an unknown permutation, based on observing $y = Π^* A x^* + w$, where $x^* \in \mathbb{R}^d$ is an unknown vector, $Π^*$ is an unknown $n \times n$ permutation matrix, and $w \in \mathbb{R}^n$ is additive Gaussian noise. We analyze the problem of permutation recovery in a random design setting in which the entries of the matrix $A$ are drawn i.i.d. from a standard Gaussian distribution, and establish sharp conditions on the SNR, sample size $n$, and dimension $d$ under which $Π^*$ is exactly and approximately recoverable. On the computational front, we show that the maximum likelihood estimate of $Π^*$ is NP-hard to compute, while also providing a polynomial time algorithm when $d =1$.

math.ST↗