Fast learning rate of non-sparse multiple kernel learning and optimal regularization strategies (Q1657947)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Fast learning rate of non-sparse multiple kernel learning and optimal regularization strategies |
scientific article |
Statements
Fast learning rate of non-sparse multiple kernel learning and optimal regularization strategies (English)
0 references
14 August 2018
0 references
The study provides a framework for a fast learning rate for multiple kernel learning (MKL) with arbitrary mixed-norm-type regularization, discussing the bound both in homogeneous as well as inhomogeneous settings. The aim of the paper is to theoretically demonstrate the superior performance of dense type MKL methods obtained within numerical experimentation.
0 references
multiple kernel learning
0 references
fast learning rate
0 references
regularization
0 references
generalization error
0 references
error bound
0 references
reproducing kernel Hilbert spaces
0 references
0 references
0 references
0 references
