English
Related papers

Related papers: Matrix denoising: Bayes-optimal estimators via low…

200 papers

Bayesian optimization has recently emerged as a popular and efficient tool for global optimization and hyperparameter tuning. Currently, the established Bayesian optimization practice requires a user-defined bounding box which is assumed to…

Machine Learning · Statistics 2015-08-18 Bobak Shahriari , Alexandre Bouchard-Côté , Nando de Freitas

We introduce a new notion of the determinant, called symmetrized determinant, for a square matrix with the entries in an associative algebra $\AA$. The monomial expansion of the symmetrized determinant is obtained from the standard…

Combinatorics · Mathematics 2007-05-23 Alexander Barvinok

Bayesian optimization is a popular and versatile approach that is well suited to solve challenging optimization problems. Their popularity comes from their effective minimization of expensive function evaluations, their capability to…

Optimization and Control · Mathematics 2026-05-14 André L. Marchildon , David W. Zingg

The vast majority of statistical theory on binary classification characterizes performance in terms of accuracy. However, accuracy is known in many cases to poorly reflect the practical consequences of classification error, most famously in…

Statistics Theory · Mathematics 2022-09-27 Shashank Singh , Justin Khim

The low-complexity assumption in linear systems can often be expressed as rank deficiency in data matrices with generalized Hankel structure. This makes it possible to denoise the data by estimating the underlying structured low-rank…

Systems and Control · Electrical Eng. & Systems 2021-11-10 Mingzhou Yin , Roy S. Smith

The matrix completion problem consists in reconstructing a matrix from a sample of entries, possibly observed with noise. A popular class of estimator, known as nuclear norm penalized estimators, are based on minimizing the sum of a data…

Statistics Theory · Mathematics 2015-04-21 Jean Lafond

We propose an efficient meta-algorithm for Bayesian estimation problems that is based on low-degree polynomials, semidefinite programming, and tensor decomposition. The algorithm is inspired by recent lower bound constructions for…

Data Structures and Algorithms · Computer Science 2017-10-04 Samuel B. Hopkins , David Steurer

The problem of estimating a high-dimensional sparse vector $\boldsymbol{\theta} \in \mathbb{R}^n$ from an observation in i.i.d. Gaussian noise is considered. The performance is measured using squared-error loss. An empirical Bayes shrinkage…

Information Theory · Computer Science 2018-12-31 Pavan Srinath , Ramji Venkataramanan

Gradient descent for matrix factorization exhibits an implicit bias toward approximately low-rank solutions. While existing theories often assume the boundedness of iterates, empirically the bias persists even with unbounded sequences. This…

Machine Learning · Computer Science 2025-11-04 Yikun Hou , Suvrit Sra , Alp Yurtsever

The Bayesian formulation of inverse problems is attractive for three primary reasons: it provides a clear modelling framework; means for uncertainty quantification; and it allows for principled learning of hyperparameters. The posterior…

Statistics Theory · Mathematics 2019-05-14 Matthew M. Dunlop , Tapio Helin , Andrew M. Stuart

Orthogonal polynomial approximations form the foundation to a set of well-established methods for uncertainty quantification known as polynomial chaos. These approximations deliver models for emulating physical systems in a variety of…

Computational Engineering, Finance, and Science · Computer Science 2022-03-23 Chun Yui Wong , Pranay Seshadri , Andrew B. Duncan , Ashley Scillitoe , Geoffrey Parks

Given a full rank matrix $X$ with more columns than rows, consider the task of estimating the pseudo inverse $X^+$ based on the pseudo inverse of a sampled subset of columns (of size at least the number of rows). We show that this is…

Machine Learning · Computer Science 2018-06-07 Michał Dereziński , Manfred K. Warmuth

Factorizing low-rank matrices has many applications in machine learning and statistics. For probabilistic models in the Bayes optimal setting, a general expression for the mutual information has been proposed using heuristic statistical…

Information Theory · Computer Science 2017-03-24 Jean Barbier , Mohamad Dia , Nicolas Macris , Florent Krzakala , Thibault Lesieur , Lenka Zdeborova

In this paper, we aim to design and analyze distributed Bayesian estimation algorithms for sensor networks. The challenges we address are to (i) derive a distributed provably-correct algorithm in the functional space of probability…

Machine Learning · Computer Science 2025-03-25 Parth Paritosh , Nikolay Atanasov , Sonia Martinez

The permanent vs. determinant problem is one of the most important problems in theoretical computer science, and is the main target of geometric complexity theory proposed by Mulmuley and Sohoni. The current best lower bound for the…

Computational Complexity · Computer Science 2015-04-02 Akihiro Yabe

We address the fundamental problem of selection under uncertainty by modeling it from the perspective of Bayesian persuasion. In our model, a decision maker with imperfect information always selects the option with the highest expected…

Computer Science and Game Theory · Computer Science 2024-10-16 Siddhartha Banerjee , Kamesh Munagala , Yiheng Shen , Kangning Wang

In a smooth semiparametric estimation problem, the marginal posterior for the parameter of interest is expected to be asymptotically normal and satisfy frequentist criteria of optimality if the model is endowed with a suitable prior. It is…

Statistics Theory · Mathematics 2012-05-30 P. J. Bickel , B. J. K. Kleijn

In nonparametric statistics, rate-optimal estimators typically balance bias and stochastic error. The recent work on overparametrization raises the question whether rate-optimal estimators exist that do not obey this trade-off. In this work…

Statistics Theory · Mathematics 2024-06-21 Alexis Derumigny , Johannes Schmidt-Hieber

When training an estimator such as a neural network for tasks like image denoising, it is often preferred to train one estimator and apply it to all noise levels. The de facto training protocol to achieve this goal is to train the estimator…

Machine Learning · Computer Science 2020-07-20 Abhiram Gnansambandam , Stanley H. Chan

Diffusion models (DMs) as generative priors have recently shown great potential for denoising tasks but lack theoretical understanding with respect to their mean square error (MSE) optimality. This paper proposes a novel denoising strategy…

Machine Learning · Computer Science 2025-03-04 Benedikt Fesl , Benedikt Böck , Florian Strasser , Michael Baur , Michael Joham , Wolfgang Utschick