2011
DOI: 10.1016/j.neucom.2010.10.012
|View full text |Cite
|
Sign up to set email alerts
|

Face recognition using kernel entropy component analysis

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

2013
2013
2022
2022

Publication Types

Select...
5
4
1

Relationship

0
10

Authors

Journals

citations
Cited by 51 publications
(16 citation statements)
references
References 6 publications
0
16
0
Order By: Relevance
“…First, we upgrade LLE algorithm with KECA-based ideas [8,9], then turn the global nonlinear problem into global linear problem in high dimensional kernel space through kernel function mapping. Assume the dimension of sample X is N, anf the probability density is p(x), then the Reny entropy is:…”
Section: Kele Algorithmmentioning
confidence: 99%
“…First, we upgrade LLE algorithm with KECA-based ideas [8,9], then turn the global nonlinear problem into global linear problem in high dimensional kernel space through kernel function mapping. Assume the dimension of sample X is N, anf the probability density is p(x), then the Reny entropy is:…”
Section: Kele Algorithmmentioning
confidence: 99%
“…Another c 2016 Information Processing Society of Japan challenge in online learning is that the support set may grow to arbitrarily large size over time, some strategies as truncation and shrinking can be applied [103]. Some other dimensionality reductions as Fisher criteria [104], [105] and kernel entropy component analysis (kernel ECA) [106], [107] are also proposed. Table 4 lists some qualitative comparison of the algorithms mentioned in this section.…”
Section: Kernel Subspace Learningmentioning
confidence: 99%
“…Compared to the widely used KPCA [26], KECA is a novel technique in the dimension reduction field [27,28]. It manages to maintain the maximum Renyi entropy of input space data set, and in contrast to KPCA maintains second-order statistics of data set maximally.…”
Section: Kecamentioning
confidence: 99%