Consistency and error analysis of Prior-Knowledge-Based Kernel Regression

Authors:
Z. Sun;Z. Zhang;H. Wang
Affiliations:
Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing 100084, PR China;Department of Automation, Tsinghua University, Beijing 100084, PR China;Department of Automation, Tsinghua University, Beijing 100084, PR China
Venue:
Neurocomputing
Year:
2011

Citing 15
Cited 1

The connection between regularization operators and support vector kernels

Neural Networks
Least Squares Support Vector Machine Classifiers

Neural Processing Letters
Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond

Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond
A tutorial on support vector regression

Statistics and Computing
Knowledge-Based Kernel Approximation

The Journal of Machine Learning Research
Learning from Examples as an Inverse Problem

The Journal of Machine Learning Research
Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples

The Journal of Machine Learning Research
Optimal Rates for the Regularized Least-Squares Algorithm

Foundations of Computational Mathematics
Incorporating prior knowledge in support vector regression

Machine Learning
Incorporating prior knowledge in support vector machines for classification: A review

Neurocomputing
Generalizing the bias term of support vector machines

IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Generalization performance of regularization networks and support vector machines via entropy numbers of compact operators

IEEE Transactions on Information Theory
Covering numbers for support vector machines

IEEE Transactions on Information Theory
Consistency of support vector machines and other regularized kernel classifiers

IEEE Transactions on Information Theory
Nonlinear Knowledge in Kernel Approximation

IEEE Transactions on Neural Networks

Estimating α-frontier technical efficiency with shape-restricted kernel quantile regression

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

Incorporating prior knowledge (PK) into learning methods is an effective means to improve learning performance. The consistency and error theories of PK-based methods, which are of great theoretical importance, are still far from well established. Concentrating on the PK-based kernel regression, this paper proposes a methodology of analyzing the consistency and error. This methodology converts the specific methods firstly to a unified optimization problem and then to a unified solution expression, and a general consistency and error analysis tool is proposed and applied. A few examples are given to illustrate the analysis procedure.