Optimal rates for regularization of statistical inverse learning problems
From MaRDI portal
inverse problemminimax convergence ratesreproducing kernel Hilbert spacespectral regularizationstatistical learning
Linear operators in reproducing-kernel Hilbert spaces (including de Branges, de Branges-Rovnyak, and other structured spaces) (47B32) Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20) Numerical solution to inverse problems in abstract spaces (65J22) Computational learning theory (68Q32)
Abstract: We consider a statistical inverse learning problem, where we observe the image of a function through a linear operator at i.i.d. random design points , superposed with an additive noise. The distribution of the design points is unknown and can be very general. We analyze simultaneously the direct (estimation of ) and the inverse (estimation of ) learning problems. In this general framework, we obtain strong and weak minimax optimal rates of convergence (as the number of observations grows large) for a large class of spectral regularization methods over regularity classes defined through appropriate source conditions. This improves on or completes previous results obtained in related settings. The optimality of the obtained rates is shown not only in the exponent in but also in the explicit dependency of the constant factor in the variance of the noise and the radius of the source condition set.
Recommendations
- Inverse statistical learning
- Convergence Rates of Spectral Regularization Methods: A Comparison between Ill-Posed Inverse Problems and Statistical Kernel Learning
- Optimal rates of convergence for nonparametric statistical inverse problems
- A unified approach to inversion problems in statistics
- Convergence analysis of Tikhonov regularization for non-linear statistical inverse problems
Cites work
- A distribution-free theory of nonparametric regression
- Approximation in learning theory
- Approximation methods for supervised learning
- Best choices for regularization parameters in learning theory: on the bias-variance problem.
- Boosting With theL2Loss
- Convergence Characteristics of Methods of Regularization Estimators for Nonlinear Operator Equations
- Convergence Rates of General Regularization Methods for Statistical Inverse Problems and Applications
- Convergence rates of kernel conjugate gradient for random design regression
- Cross-validation based adaptation for regularization operators in learning theory
- DISCRETIZATION ERROR ANALYSIS FOR TIKHONOV REGULARIZATION
- Fréchet derivatives of the power function
- Geometry of linear ill-posed problems in variable Hilbert scales
- scientific article; zbMATH DE number 3824308 (Why is no real title available?)
- scientific article; zbMATH DE number 3907465 (Why is no real title available?)
- scientific article; zbMATH DE number 45848 (Why is no real title available?)
- scientific article; zbMATH DE number 3605460 (Why is no real title available?)
- scientific article; zbMATH DE number 4000257 (Why is no real title available?)
- scientific article; zbMATH DE number 936298 (Why is no real title available?)
- scientific article; zbMATH DE number 967931 (Why is no real title available?)
- Introduction to nonparametric estimation
- Inverse statistical learning
- Learning from examples as an inverse problem
- Learning theory estimates via integral operators and their approximations
- Minimax fast rates for discriminant analysis with errors in variables
- On early stopping in gradient descent learning
- On regularization algorithms in learning theory
- Optimal learning rates for least squares regularized regression with unbounded sampling
- Optimal rates for the regularized least-squares algorithm
- Regularization in kernel learning
- Shannon sampling. II: Connections to learning theory
- Spectral Algorithms for Supervised Learning
- Statistical consistency of kernel canonical correlation analysis
- Support Vector Machines
Cited in
(71)- Kernel conjugate gradient methods with random projections
- An elementary analysis of ridge regression with random design
- Bayesian frequentist bounds for machine learning and system identification
- From inexact optimization to learning via gradient concentration
- Lower bounds for invariant statistical models with applications to principal component analysis
- Rates of convergence of randomized Kaczmarz algorithms in Hilbert spaces
- Convergence of regularization methods with filter functions for a regularization parameter chosen with GSURE and mildly ill-posed inverse problems
- Learning rates for the kernel regularized regression with a differentiable strongly convex loss
- Tikhonov regularization with oversmoothing penalty for nonlinear statistical inverse problems
- Convergence analysis of Tikhonov regularization for non-linear statistical inverse problems
- The empirical process of residuals from an inverse regression
- Concentration of weakly dependent Banach-valued sums and applications to statistical learning methods
- Optimal rate of the regularized regression learning algorithm
- scientific article; zbMATH DE number 6671876 (Why is no real title available?)
- Kernel regression, minimax rates and effective dimensionality: beyond the regular case
- Convergence Rates of Spectral Regularization Methods: A Comparison between Ill-Posed Inverse Problems and Statistical Kernel Learning
- Radial basis function regularization for linear inverse problems with random noise
- Nyström subsampling method for coefficient-based regularized regression
- Convergences of regularized algorithms and stochastic gradient methods with random projections
- Distributed minimum error entropy algorithms
- Analyzing the discrepancy principle for kernelized spectral filter learning algorithms
- Error analysis of the kernel regularized regression based on refined convex losses and RKBSs
- Regularization: From Inverse Problems to Large-Scale Machine Learning
- Two-Layer Neural Networks with Values in a Banach Space
- Distributed spectral pairwise ranking algorithms
- Shearlet-based regularization in statistical inverse learning with an application to x-ray tomography
- scientific article; zbMATH DE number 7306853 (Why is no real title available?)
- Sobolev norm learning rates for regularized least-squares algorithms
- On the improved rates of convergence for Matérn-type kernel ridge regression with application to calibration of computer models
- scientific article; zbMATH DE number 7415114 (Why is no real title available?)
- Online regularized pairwise learning with least squares loss
- Convergence analysis of distributed multi-penalty regularized pairwise learning
- Convergence Rates of General Regularization Methods for Statistical Inverse Problems and Applications
- Optimal indirect estimation for linear inverse problems with discretely sampled functional data
- A note on the prediction error of principal component regression in high dimensions
- Mini-workshop: Mathematical foundations of robust and generalizable learning. Abstracts from the mini-workshop held October 2--8, 2022
- Convergence Rates for Learning Linear Operators from Noisy Data
- Convex regularization in statistical inverse learning problems
- Optimality of regularized least squares ranking with imperfect kernels
- Inverse learning in Hilbert scales
- Optimal learning rates for least squares regularized regression with unbounded sampling
- Sketching with Spherical Designs for Noisy Data Fitting on Spheres
- Nonlinear Tikhonov regularization in Hilbert scales for inverse learning
- Learning particle swarming models from data with Gaussian processes
- Iterative kernel regression with preconditioning
- Spectral algorithms for functional linear regression
- How many neurons do we need? A refined analysis for shallow networks trained with gradient descent
- Least squares approximations in linear statistical inverse learning problems
- Optimality of robust online learning
- Adaptive parameter selection for kernel ridge regression
- Learning sparsity-promoting regularizers for linear inverse problems
- Convex analysis method for distributed learning algorithm
- Robust kernel-based gradient descent with random features
- Optimal rates of convergence for nonparametric statistical inverse problems
- Improved convergence rate of nested simulation with LSE on sieve
- Nonlinear meta-learning can guarantee faster rates
- Frontiers of statistics and machine learning. Abstracts from the workshop held March 23--28, 2025
- Regularization with optimal space-time priors
- Convergence rates of regularized Huber regression under weak moment conditions
- Distributed learning with discretely observed functional data
- Functional linear and single-index models: a unified approach via Gaussian Stein identity
- Criticality measure-based error estimates for infinite dimensional optimization
- A Lipschitz spaces view of infinitely wide shallow neural networks
- Gradient-based nonlinear inverse learning
- Learning theory of regularized Huber regression
- Sobolev norm inconsistency of kernel interpolation
- Optimal Plug-in Gaussian Processes for Modeling Derivatives
- Statistical learning with sublinear regret of propagator models
- Learning Lévy density via adaptive RKHS regression with bi-level optimization
- Sampling theorems for inverse problems on Riemannian manifolds
- Construction and Monte Carlo estimation of wavelet frames generated by a reproducing kernel
This page was built for publication: Optimal rates for regularization of statistical inverse learning problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q667648)
