Geometric inference for general high-dimensional linear inverse problems
From MaRDI portal
Abstract: This paper presents a unified geometric framework for the statistical analysis of a general ill-posed linear inverse model which includes as special cases noisy compressed sensing, sign vector recovery, trace regression, orthogonal matrix estimation, and noisy matrix completion. We propose computationally feasible convex programs for statistical inference including estimation, confidence intervals and hypothesis testing. A theoretical framework is developed to characterize the local estimation rate of convergence and to provide statistical inference guarantees. Our results are built based on the local conic geometry and duality. The difficulty of statistical inference is captured by the geometric characterization of the local tangent cone through the Gaussian width and Sudakov minoration estimate.
Recommendations
- The convex geometry of linear inverse problems
- Estimation in high dimensions: a geometric perspective
- Fast global convergence of gradient methods for high-dimensional statistical recovery
- A unified framework for high-dimensional analysis of M-estimators with decomposable regularizers
- High-dimensional estimation with geometric constraints
Cites work
- A constrained \(\ell _{1}\) minimization approach to sparse precision matrix estimation
- A statistical perspective on ill-posed inverse problems (with discussion)
- A unified framework for high-dimensional analysis of M-estimators with decomposable regularizers
- An adaptation theory for nonparametric confidence intervals
- Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
- Estimation of high-dimensional low-rank matrices
- Exact matrix completion via convex optimization
- Geometric inference for general high-dimensional linear inverse problems
- Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization
- How well can we estimate a sparse vector?
- scientific article; zbMATH DE number 3787949 (Why is no real title available?)
- scientific article; zbMATH DE number 49190 (Why is no real title available?)
- scientific article; zbMATH DE number 194093 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- Inferring Rankings Using Constrained Sensing
- Information-theoretic determination of minimax rates of convergence
- Living on the edge: phase transitions in convex programs with random data
- Majorizing measures: The generic chaining
- Matrix completion via max-norm constrained optimization
- Matrix estimation by universal singular value thresholding
- Minimax estimation of linear functionals over nonconvex parameter spaces.
- NP is as easy as detecting unique solutions
- Nuclear-norm penalization and optimal rates for noisy low-rank matrix completion
- On asymptotically optimal confidence regions and tests for high-dimensional models
- On complexity of unconstrained hyperbolic 0--1 programming problems
- Orthogonal Procrustes rotation for two or more matrices
- Probability of unique integer solution to a system of linear equations
- Procrustes Problems
- Rate minimaxity of the Lasso and Dantzig selector for the \(l_{q}\) loss in \(l_{r}\) balls
- Robust principal component analysis?
- Simultaneous analysis of Lasso and Dantzig selector
- Speed of estimation in positron emission tomography and related inverse problems
- Statistical estimation and optimal recovery
- Statistical significance in high-dimensional linear models
- Statistics for high-dimensional data. Methods, theory and applications.
- The convex geometry of linear inverse problems
- The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
- The sizes of compact subsets of Hilbert space and continuity of Gaussian processes
- Tight Oracle Inequalities for Low-Rank Matrix Recovery From a Minimal Number of Noisy Random Measurements
- Volume Ratio, Sparsity, and Minimaxity Under Unitarily Invariant Norms
- Von Neumann entropy penalization and low-rank matrix estimation
Cited in
(14)- Linear regression with sparsely permuted data
- The convex geometry of linear inverse problems
- Rate optimal estimation and confidence intervals for high-dimensional regression with missing covariates
- On the robustness of minimum norm interpolators and regularized empirical risk minimizers
- Inference for low-rank tensors -- no need to debias
- Estimation in high dimensions: a geometric perspective
- Geometric inference for general high-dimensional linear inverse problems
- High-dimensional estimation with geometric constraints
- Weighted message passing and minimum energy flow for heterogeneous stochastic block models with side information
- scientific article; zbMATH DE number 7370563 (Why is no real title available?)
- Plug in estimation in high dimensional linear inverse problems a rigorous analysis
- Noisy linear inverse problems under convex constraints: exact risk asymptotics in high dimensions
- Statistical inference for matrix-vector linear regression without debiasing under Kronecker covariance structure
- Statistical inference in tensor completion: optimal uncertainty quantification and statistical-to-computational gaps
This page was built for publication: Geometric inference for general high-dimensional linear inverse problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q309721)
