2011
DOI: 10.1007/s10115-011-0404-6
|View full text |Cite
|
Sign up to set email alerts
|

A survey of the state of the art in learning the kernels

Abstract: In recent years, the machine learning community has witnessed a tremendous growth in the development of kernel-based learning algorithms. However, the performance of this class of algorithms greatly depends on the choice of the kernel function. Kernel function implicitly represents the inner product between a pair of points of a dataset in a higher dimensional space. This inner product amounts to the similarity between points and provides a solid foundation for nonlinear analysis in kernel-based learning algor… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
16
0

Year Published

2012
2012
2023
2023

Publication Types

Select...
6
2

Relationship

0
8

Authors

Journals

citations
Cited by 39 publications
(16 citation statements)
references
References 75 publications
0
16
0
Order By: Relevance
“…e interested reader may refer to the recent survey on kernel learning by Abbasnejad et al [2012]. ese approaches are thus powerful but limited to the transductive setting: the resulting kernel is difficult to use on new data.…”
Section: Related Topicsmentioning
confidence: 99%
“…e interested reader may refer to the recent survey on kernel learning by Abbasnejad et al [2012]. ese approaches are thus powerful but limited to the transductive setting: the resulting kernel is difficult to use on new data.…”
Section: Related Topicsmentioning
confidence: 99%
“…It is another way to obtain the kernel that the weighting coefficients of the basic kernels are optimized [28][29][30][31].…”
Section: Mk-svm With the Unweighted Sum Of Baisc Kernelsmentioning
confidence: 99%
“…Support vector machine (SVM) is a state-of-art machine learning approach [26,27]. Multiple kernel support vector machine (MK-SVM) is a popular topic in kernel methods [28][29][30][31]42,44,45]. It aims at learning the optimal kernel function by optimizing the combination of multiple heterogeneous basic kernels [31,45], multiple basic kernels with different feature subsets [28] or hyperparameters [44].…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…As for the codebook, we set the number of clusters to be 1024 in k-means, and randomly select 8 * 10 4 features from the whole training feature set to generate codebook for each data set. Then the ScSPM [42] method is used for coding, in which the spatial block number on each level of the pyramid is set as [1,2,4], the weight for features on each level is set as . We use the matlab toolbox of BPDN-homotopy provided on the webpage http://users.ece.gatech.edu/%7Esasif/homotopy/ (The same is also provided on http://www.eecs.berkeley.edu/∼yang/software/l1benchmark/).…”
Section: A Experimental Setupmentioning
confidence: 99%