中文
相关论文

相关论文: Lower Bounds and a Near-Optimal Shrinkage Estimato…

200 篇论文

We study randomized sketching methods for approximately solving least-squares problem with a general convex constraint. The quality of a least-squares approximation can be assessed in different ways: either in terms of the value of the…

最优化与控制 · 数学 2014-11-04 Mert Pilanci , Martin J. Wainwright

Consider the problem of estimating the mean of a Gaussian random vector when the mean vector is assumed to be in a given convex set. The most natural solution is to take the Euclidean projection of the data vector on to this convex set; in…

统计理论 · 数学 2014-11-21 Sourav Chatterjee

We consider statistical and algorithmic aspects of solving large-scale least-squares (LS) problems using randomized sketching algorithms. Prior results show that, from an \emph{algorithmic perspective}, when using sketching matrices…

机器学习 · 统计学 2015-05-26 Garvesh Raskutti , Michael Mahoney

Algorithmic Gaussianization is a phenomenon that can arise when using randomized sketching or sampling methods to produce smaller representations of large datasets: For certain tasks, these sketched representations have been observed to…

机器学习 · 计算机科学 2023-07-28 Michał Dereziński

Constrained least squares problems arise in many applications. Their memory and computation costs are expensive in practice involving high-dimensional input data. We employ the so-called "sketching" strategy to project the least squares…

最优化与控制 · 数学 2021-09-07 Ke Chen , Ruhui Jin

We consider distributed optimization methods for problems where forming the Hessian is computationally challenging and communication is a significant bottleneck. We leverage randomized sketches for reducing the problem dimensions as well as…

最优化与控制 · 数学 2022-03-21 Burak Bartan , Mert Pilanci

The Count-Min sketch is an important and well-studied data summarization method. It allows one to estimate the count of any item in a stream using a small, fixed size data sketch. However, the accuracy of the sketch depends on…

数据结构与算法 · 计算机科学 2018-11-13 Daniel Ting

Matrix sketching is a powerful tool for reducing the size of large data matrices. Yet there are fundamental limitations to this size reduction when we want to recover an accurate estimator for a task such as least square regression. We show…

数据结构与算法 · 计算机科学 2024-05-10 Sachin Garg , Kevin Tan , Michał Dereziński

In linear regression we wish to estimate the optimum linear least squares predictor for a distribution over $d$-dimensional input points and real-valued responses, based on a small sample. Under standard random design analysis, where the…

机器学习 · 统计学 2022-06-08 Michał Dereziński , Manfred K. Warmuth , Daniel Hsu

We consider statistical as well as algorithmic aspects of solving large-scale least-squares (LS) problems using randomized sketching algorithms. For a LS problem with input data $(X, Y) \in \mathbb{R}^{n \times p} \times \mathbb{R}^n$,…

机器学习 · 统计学 2015-08-26 Garvesh Raskutti , Michael Mahoney

We propose a Randomised Subspace Gauss-Newton (R-SGN) algorithm for solving nonlinear least-squares optimization problems, that uses a sketched Jacobian of the residual in the variable domain and solves a reduced linear least-squares on…

最优化与控制 · 数学 2022-11-11 Coralia Cartis , Jaroslav Fowkes , Zhen Shao

We consider a least squares regression problem where the data has been generated from a linear model, and we are interested to learn the unknown regression parameters. We consider "sketch-and-solve" methods that randomly project the data…

统计理论 · 数学 2019-10-08 Edgar Dobriban , Sifan Liu

Researchers may perform regressions using a sketch of data of size $m$ instead of the full sample of size $n$ for a variety of reasons. This paper considers the case when the regression errors do not have constant variance and…

机器学习 · 统计学 2022-06-23 Sokbae Lee , Serena Ng

Random projections or sketching are widely used in many algorithmic and learning contexts. Here we study the performance of iterative Hessian sketch for least-squares problems. By leveraging and extending recent results from random matrix…

最优化与控制 · 数学 2020-10-26 Jonathan Lacotte , Sifan Liu , Edgar Dobriban , Mert Pilanci

A lower bound on the minimum mean-squared error (MSE) in a Bayesian estimation problem is proposed in this paper. This bound utilizes a well-known connection to the deterministic estimation setting. Using the prior distribution, the bias…

信息论 · 计算机科学 2009-05-27 Zvika Ben-Haim , Yonina C. Eldar

We present a stochastic inexact Gauss-Newton method for the solution of nonlinear least-squares. To reduce the computational cost with respect to the classical method, at each iteration the proposed algorithm approximately minimizes the…

最优化与控制 · 数学 2025-06-05 Stefania Bellavia , Greta Malaspina , Benedetta Morini

A methodology for using random sketching in the context of model order reduction for high-dimensional parameter-dependent systems of equations was introduced in [Balabanov and Nouy 2019, Part I]. Following this framework, we here construct…

数值分析 · 数学 2022-03-25 Oleg Balabanov , Anthony Nouy

We propose a general random subspace framework for unconstrained nonconvex optimization problems that requires a weak probabilistic assumption on the subspace gradient, which we show to be satisfied by various random matrix ensembles, such…

最优化与控制 · 数学 2022-11-21 Coralia Cartis , Jaroslav Fowkes , Zhen Shao

Overparametrization often helps improve the generalization performance. This paper presents a dual view of overparametrization suggesting that downsampling may also help generalize. Focusing on the proportional regime $m\asymp n \asymp p$,…

统计理论 · 数学 2023-10-17 Xin Chen , Yicheng Zeng , Siyue Yang , Qiang Sun

Linear Least Squares is a very well known technique for parameter estimation, which is used even when sub-optimal, because of its very low computational requirements and the fact that exact knowledge of the noise statistics is not required.…

统计理论 · 数学 2018-10-16 Michael Krikheli , Amir Leshem
‹ 上一页 1 2 3 10 下一页 ›