ROOT-SGD solves convex optimization problems with optimal nonasymptotic and near-optimal asymptotic performance.
problem Solving strongly convex and smooth unconstrained optimization problems using stochastic first-order algorithms.
method ROOT-SGD: Recursive One-Over-T SGD, averaging past stochastic gradients.
result Achieves state-of-the-art performance in both nonasymptotic and asymptotic senses.
New method estimates gradients accurately with sharp bounds.
problem Accurate gradient estimation in regression problems.
method Nearest-neighbor based pointwise estimate of gradients.
result Sharp nonasymptotic bounds for gradient estimation.
New confidence intervals improve treatment effect estimation in randomized experiments.
problem Improving confidence intervals for treatment effects in randomized experiments.
method Systematic exploitation of negative dependence or variance adaptivity.
result Achieved nonasymptotic confidence intervals with the same effective sample size as asymptotic ones.
SGHMC improves sampling and optimization under local conditions.
problem Nonconvex optimization and sampling under local conditions.
method Nonasymptotic analysis of SGHMC convergence.
result SGHMC provides high-precision results uniformly in iterations.
We establish the first nonasymptotic error bounds for Kaplan-Meier-based nearest neighbor and kernel survival probability estimators where feature vectors reside in metric spaces. Our bounds imply rates of strong consistency for these nonparametric estimators and, up to a log factor, match an existing lower bound for c…
Phase retrieval refers to the problem of recovering real- or complex-valued vectors from magnitude measurements. The best-known algorithms for this problem are iterative in nature and rely on so-called spectral initializers that provide accurate initialization vectors. We propose a novel class of estimators suitable fo…
The paper improves nonparametric confidence bands for band-limited functions.
problem Constructing nonparametric simultaneous confidence bands with nonasymptotic and distribition-free guarantees.
method Based on Paley-Wiener reproducing kernel Hilbert spaces, the paper relaxes assumptions, improves noise estimation, and tightens constraints.
result Enhanced confidence bands with improved efficiency and tighter constraints.
This paper tightens the law of the iterated logarithm for empirical KL_inf, applicable to unbounded data.
problem Developing nonasymptotic concentration bounds for empirical KL_inf with optimal constants and rates.
method Presenting a tight law of the iterated logarithm for empirical KL_inf, applicable to unbounded data.
result A tight law of the iterated logarithm for empirical KL_inf, applicable to unbounded data.
Optimized AIS scheme reduces bias and MSE for general proposals.
problem Performing Monte Carlo integration with general proposals.
method Global optimization of χ²-divergence using stochastic gradient Langevin dynamics.
result Explicit theoretical guarantees for uniform-in-time MSE reduction.
FIEM accelerates EM for large datasets with nonasymptotic convergence bounds.
problem Efficiently optimizing large datasets using EM framework.
method FIEM recasts EM in Stochastic Approximation framework and provides nonasymptotic convergence bounds.
result Nonasymptotic bounds for convergence in expectation as a function of n and $\kmax$. We investigate the relationship between the structure of a discrete graphical model and the support of the inverse of a generalized covariance matrix. We show that for certain graph structures, the support of the inverse covariance matrix of indicator variables on the vertices of a graph reflects the conditional indepe…
This paper introduces time-uniform CLT-based confidence intervals for statistical inference.
problem Developing valid statistical inference methods for sequential data.
method Time-uniform central limit theory and strong invariance principles.
result Asymptotic confidence sequences (CSs) that are uniformly valid over time.
In this work, we study the problem of reconstructing shapes from simple nonasymptotic densities measured only along shape boundaries. The particular density we study is also known as the integral area invariant and corresponds to the area of a disk centered on the boundary that is also inside the shape. It is easy to s…
We propose a new algorithm---Stochastic Proximal Langevin Algorithm (SPLA)---for sampling from a log concave distribution. Our method is a generalization of the Langevin algorithm to potentials expressed as the sum of one stochastic smooth term and multiple stochastic nonsmooth terms. In each iteration, our splitting t…
Gradient descent optimally trains RNNs without overparameterization.
problem Training recurrent neural networks (RNNs) with gradient descent.
method Nonasymptotic analysis of gradient descent for RNNs with diagonal weight matrices.
result Gradient descent can achieve optimality in RNNs with a network size scaling logarithmically with the number of samples.
Stochastic Gradient Langevin Dynamics (SGLD) is a popular variant of Stochastic Gradient Descent, where properly scaled isotropic Gaussian noise is added to an unbiased estimate of the gradient at each iteration. This modest change allows SGLD to escape local minima and suffices to guarantee asymptotic convergence to g…
The paper creates nonparametric confidence bands for band-limited functions.
problem Estimating confidence bands for band-limited functions with finite samples and unknown noise.
method Uses Paley-Wiener reproducing kernel Hilbert spaces and gradient-perturbation methods.
result Non-asymptotic guarantees for confidence regions without assuming a parametric model.
Study tests whether trade-off functions are above or below benchmarks using finite samples.
problem Testing trade-off functions between unknown distributions.
method Identifies a condition for nontrivial testing, constructs a test with error guarantees, and inverts the test for confidence bands.
result Finite-sample testing is possible under specific structural assumptions about rejection regions.
New method estimates precision matrices without models, achieving dense, consistent, and model-free properties.
problem Lack of methods that are dense, consistent, and model-free for precision matrix estimation.
method General class of estimators that unify dense, consistent, and model-free properties within a nonasymptotic framework.
result Ridgeless regression exhibits the double descent phenomenon, establishing a precision matrix analogue to linear regression's double descent.
We study the concentration of random kernel matrices around their mean. We derive nonasymptotic exponential concentration inequalities for Lipschitz kernels assuming that the data points are independent draws from a class of multivariate distributions on Rd, including the strongly log-concave distributions u…
Study on signal detection in sparse additive models with nonasymptotic minimax rates.
problem Signal detection in sparse additive models.
method Nonasymptotic minimax analysis of signal detection in sparse additive models.
result Established minimax separation rate for signal detection.
Paper improves confidence intervals and variance estimation for deep learning models.
problem Improving confidence intervals and variance estimation in deep learning models.
method Residual-based framework for conditional variance estimation; robust bootstrap procedure for confidence intervals.
result First non-asymptotic bounds for variance estimation using ReLU networks.
Sampling from various kinds of distributions is an issue of paramount importance in statistics since it is often the key ingredient for constructing estimators, test procedures or confidence intervals. In many situations, the exact sampling from a given distribution is impossible or computationally expensive and, there…
New framework robustifies loss functions with quantiles for outlier resistance.
problem Widespread outliers in big data affect statistical estimation and inference.
method Introduces a framework connecting to trimming, scalable algorithms, and new techniques.
result Robust estimators achieve minimax rate optimality in regression, classification, and neural networks.
New theory explains how overparametrized neural networks generalize well without bias-variance trade-off.
problem Overparametrized neural networks generalize well despite classical bias-variance trade-off.
method Nonasymptotic generalization theory for two-layer neural networks with ReLU activation, incorporating scaled variation regularization.
result Prediction bounds for all network widths reproduce the double descent phenomenon, and overparametrized models are nearly minimax optimal.
Exact minimax risk derived for linear prediction with sample covariance analysis.
problem Understanding the minimax risk in linear prediction under various covariate distributions.
method Exact minimax risk analysis, leveraging statistical leverage scores and PAC-Bayes techniques.
result The minimax risk is of order d/(n−d+1) for any covariate distribution, nearly matching the risk for Gaussian design. Study shows generative priors improve rank-one matrix recovery with optimal sample complexity.
problem Recovering a rank-one signal matrix from noisy data with additional prior information.
method Analysis of a nonlinear least squares objective with a favorable global optimization landscape.
result Established optimal sample complexity for generative priors in rank-one matrix recovery.
Optimization geometrodynamics simplifies adaptive optimizer dynamics.
problem Hidden states in adaptive optimizers complicate gradient-based learning.
method Develops a variational theory to eliminate hidden states and compose across hierarchies.
result Yields interaction curvature that integrates to finite contrasts.
Develops a new algorithm for estimating model parameters using interacting particle systems.
problem Estimating parameters of latent variable models.
method Interacting Particle Langevin Algorithm (IPLA) based on Langevin diffusion.
result Nonasymptotic optimisation error bounds for the estimator.
This research provides theoretical guarantees for hyperparameter estimation in complex network dynamical systems.
problem Theoretical guarantees for hyperparameter estimation in large, inhomogeneous complex network dynamical systems.
method Formulating the system's evolution in a measure transport perspective, proposing a theoretical framework for estimating hyperparameters with mean-type observations.
result A nonasymptotic bound for the deviation of hyperparameter estimates in inhomogeneous complex network dynamical systems with respect to network population size.
The paper develops tests for comparing means in high dimensions with unknown covariance.
problem Testing if the mean of a high-dimensional distribution is close to zero or different from another.
method Develops nonasymptotic tests using concentration inequalities and operator norms.
result Obtains bounds on the minimal separation distance for controlling Type I and Type II errors.
We study the problem of robustly estimating the posterior distribution for the setting where observed data can be contaminated with potentially adversarial outliers. We propose Rob-ULA, a robust variant of the Unadjusted Langevin Algorithm (ULA), and provide a finite-sample analysis of its sampling distribution. In par…
We consider the problem of high-dimensional Ising (graphical) model selection. We propose a simple algorithm for structure estimation based on the thresholding of the empirical conditional variation distances. We introduce a novel criterion for tractable graph families, where this method is efficient, based on the pres…
Modern geometric measure theory, developed largely to solve the Plateau problem, has generated a great deal of technical machinery which is unfortunately regarded as inaccessible by outsiders. Some of its tools (e.g., flat norm distance and decomposition in generalized surface space) hold interest from a theoretical pe…
The paper provides bounds for regression schemes using nonstationary training samples.
problem Developing confidence intervals for nonparametric regression with nonstationary data.
method The approach involves Rademacher and Vapnik-Chervonenkis theories to analyze the cost and optimality of regression schemes.
result The paper establishes nonasymptotic bounds for regression schemes and optimality in L2-distance. In this paper, we are concerned with a non-asymptotic analysis of sampling algorithms used in nonconvex optimization. In particular, we obtain non-asymptotic estimates in Wasserstein-1 and Wasserstein-2 distances for a popular class of algorithms called Stochastic Gradient Langevin Dynamics (SGLD). In addition, the afo…
We take a Hamiltonian-based perspective to generalize Nesterov's accelerated gradient descent and Polyak's heavy ball method to a broad class of momentum methods in the setting of (possibly) constrained minimization in Euclidean and non-Euclidean normed vector spaces. Our perspective leads to a generic and unifying non…
Langevin Monte Carlo (LMC) is an iterative algorithm used to generate samples from a distribution that is known only up to a normalizing constant. The nonasymptotic dependence of its mixing time on the dimension and target accuracy is understood mainly in the setting of smooth (gradient-Lipschitz) log-densities, a seri…
Kronheimer and Mrowka introduced a new knot invariant, called s♯, which is a gauge theoretic analogue of Rasmussen's s invariant. In this article, we compute Kronheimer and Mrowka's invariant for some classes of knots, including algebraic knots and the connected sums of quasi-positive knots with non-trivial r…
The paper explains how nearest neighbor methods succeed in prediction.
problem Explaining the success of nearest neighbor methods in prediction.
method The paper covers both theoretical and practical aspects of nearest neighbor methods, including statistical guarantees and practical algorithms.
result The paper provides nonasymptotic statistical guarantees and practical algorithms for nearest neighbor methods.
In this paper, we explore a general Aggregated Gradient Langevin Dynamics framework (AGLD) for the Markov Chain Monte Carlo (MCMC) sampling. We investigate the nonasymptotic convergence of AGLD with a unified analysis for different data accessing (e.g. random access, cyclic access and random reshuffle) and snapshot upd…
Efficiently transforms samples from various statistical models.
problem Approximately transforming samples from one statistical model to another without knowing the source model's parameters.
method Constructs computationally efficient procedures to reduce uniform, Erlang, and Laplace models to general target families.
result Establishes nonasymptotic reductions between canonical high-dimensional problems, such as mixtures of experts, phase retrieval, and signal denoising.
Deep networks can perfectly classify two low-dimensional manifolds on a sphere with large depth and width.
problem Binary classification of two low-dimensional submanifolds on a sphere.
method Analysis of a deep fully-connected neural network trained to separate two submanifolds of the unit sphere.
result Randomly-initialized gradient descent can perfectly classify the two manifolds with high probability when the network depth is large relative to certain geometric and statistical properties of the data.
Sharp inequality found on three-balls for fourth order Sobolev traces.
problem Fourth order Sobolev trace inequality on three-balls.
method Established through equivalence to a third order Sobolev inequality on two-spheres.
result Sharp fourth order Sobolev trace inequality on three-balls.
Sharp-MAML improves MAML by reducing saddle points in few-shot learning.
problem Challenges in optimizing MAML due to complex loss landscape.
method Sharpness-aware minimization applied to MAML.
result Sharp-MAML and its variant outperform plain MAML on few-shot learning tasks.
Sharp estimates for mean curvature flow of graphs are shown and examples are given to illustrate why these are sharp. The estimates improves earlier (non-sharp) estimates of Klaus Ecker and Gerhard Huisken.
ConquerNet smooths quantile regression for deep learning with minimax guarantees.
problem Optimization challenges in quantile regression for deep models.
method ConquerNet uses convolution-smoothed quantile ReLU neural networks.
result ConquerNet provides minimax guarantees and outperforms standard quantile neural networks.
Sharpe ratio (sometimes also referred to as information ratio) is widely used in asset management to compare and benchmark funds and asset managers. It computes the ratio of the (excess) net return over the strategy standard deviation. However, the elements to compute the Sharpe ratio, namely, the expected returns and …