Error bounds for \(l^p\)-norm multiple kernel learning with least square loss (Q448851): Difference between revisions
From MaRDI portal
Created a new Item |
ReferenceBot (talk | contribs) Changed an Item |
||
(6 intermediate revisions by 6 users not shown) | |||
Property / review text | |||
Summary: The problem of learning the kernel function with linear combinations of multiple kernels has attracted considerable attention recently in machine learning. Specially, by imposing an \(l^p\)-norm penalty on the kernel combination coefficient, multiple kernel learning (MKL) was proved useful and effective for theoretical analysis and practical applications. In this paper, we present a theoretical analysis on the approximation error and learning ability of the \(l^p\)-norm MKL. Our analysis shows explicit learning rates for \(l^p\)-norm MKL and demonstrates some notable advantages compared with traditional kernel-based learning algorithms where the kernel is fixed. | |||
Property / review text: Summary: The problem of learning the kernel function with linear combinations of multiple kernels has attracted considerable attention recently in machine learning. Specially, by imposing an \(l^p\)-norm penalty on the kernel combination coefficient, multiple kernel learning (MKL) was proved useful and effective for theoretical analysis and practical applications. In this paper, we present a theoretical analysis on the approximation error and learning ability of the \(l^p\)-norm MKL. Our analysis shows explicit learning rates for \(l^p\)-norm MKL and demonstrates some notable advantages compared with traditional kernel-based learning algorithms where the kernel is fixed. / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 68T05 / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 62H30 / rank | |||
Normal rank | |||
Property / zbMATH DE Number | |||
Property / zbMATH DE Number: 6078833 / rank | |||
Normal rank | |||
Property / Wikidata QID | |||
Property / Wikidata QID: Q58697082 / rank | |||
Normal rank | |||
Property / describes a project that uses | |||
Property / describes a project that uses: ElemStatLearn / rank | |||
Normal rank | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Property / full work available at URL | |||
Property / full work available at URL: https://doi.org/10.1155/2012/915920 / rank | |||
Normal rank | |||
Property / OpenAlex ID | |||
Property / OpenAlex ID: W1985786749 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q5396629 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3093181 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3093293 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Theory of Reproducing Kernels / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: 10.1162/153244302760200704 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: On the mathematical foundations of learning / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Consistency analysis of spectral regularization algorithms / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3174075 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Learning rates of least-square regularized regression / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: The elements of statistical learning. Data mining, inference, and prediction / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Multi-kernel regularized classifiers / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Model Selection and Estimation in Regression with Grouped Variables / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Weak convergence and empirical processes. With applications to statistics / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Learning theory estimates via integral operators and their approximations / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Shannon sampling. II: Connections to learning theory / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Error bounds for learning the kernel / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Capacity of reproducing kernel spaces in learning theory / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: A note on application of integral operator in learning theory / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Graph-Based Semi-Supervised Learning and Spectral Kernel Design / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Statistical Learning Theory: Models, Concepts, and Results / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q2880875 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Sparsity in penalized empirical risk minimization / rank | |||
Normal rank | |||
links / mardi / name | links / mardi / name | ||
Latest revision as of 16:31, 5 July 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Error bounds for \(l^p\)-norm multiple kernel learning with least square loss |
scientific article |
Statements
Error bounds for \(l^p\)-norm multiple kernel learning with least square loss (English)
0 references
7 September 2012
0 references
Summary: The problem of learning the kernel function with linear combinations of multiple kernels has attracted considerable attention recently in machine learning. Specially, by imposing an \(l^p\)-norm penalty on the kernel combination coefficient, multiple kernel learning (MKL) was proved useful and effective for theoretical analysis and practical applications. In this paper, we present a theoretical analysis on the approximation error and learning ability of the \(l^p\)-norm MKL. Our analysis shows explicit learning rates for \(l^p\)-norm MKL and demonstrates some notable advantages compared with traditional kernel-based learning algorithms where the kernel is fixed.
0 references