Approximate minimization of the regularized expected error over kernel models

DOI10.1287/MOOR.1080.0317MaRDI QIDQ3168992zbMATH OpenOpenAlexFDO

Authors Věra Kůrková, Marcello Sanguineti

Publication date 27 April 2011

Published in Mathematics of Operations Research (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1287/moor.1080.0317

zbMATH Keywords

convex functionalsexpected errorkernel methodsmodel complexityrates of convergencesuboptimal solutions

Mathematics Subject Classification ID

Rate of convergence, degree of approximation (41A25) Applications of variational problems in infinite-dimensional spaces to the sciences (58E50) Computational learning theory (68Q32) Programming in abstract spaces (90C48)

Recommendations

• Learning with generalization capability by kernel methods of bounded complexity
• Regularization in kernel learning
• Learning rates of least-square regularized regression
• Optimal rate of the regularized regression learning algorithm
• Learning rates for l^1-regularized kernel classifiers

Cited in

(17)

• Estimates of the approximation error using Rademacher complexity: Learning vector-valued functions
• On spectral windows in supervised learning from data
• On asymptotic properties of hyperparameter estimators for kernel-based regularization methods
• Kernel estimation under linear-exponential loss
• A meta-learning approach to the regularized learning -- case study: blood glucose prediction
• Approximate dynamic programming for stochastic \(N\)-stage optimization with application to optimal consumption under uncertainty
• Can dictionary-based computational models outperform the best linear ones?
• Suboptimal Policies for Stochastic N-Stage Optimization: Accuracy Analysis and a Case Study from Optimal Consumption
• On expected error of randomized Nyström kernel regression
• Learning with boundary conditions
• Functional optimization by variable-basis approximation schemes
• Suboptimal solutions to dynamic optimization problems via approximations of the policy functions
• Data-driven performance metrics for neural network learning
• New insights into Witsenhausen's counterexample
• Ideal regularization for learning kernels from labels
• Approximate kernel competitive learning
• Estimates of variation with respect to a set and applications to optimization problems

Describes a project that uses

Uses Software

• PDCO

This page was built for publication: Approximate minimization of the regularized expected error over kernel models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3168992)