2019
DOI: 10.1109/access.2019.2904727
|View full text |Cite
|
Sign up to set email alerts
|

Unsupervised Robust Multiple Kernel Learning via Extracting Local and Global Noises

Abstract: Kernel-based clustering methods can capture the non-linear structure and identify arbitrarily shaped clusters, so they have been widely used in machine learning tasks. Since the performance of kernel methods critically depends on the choices of kernels, multiple kernel learning methods are proposed to alleviate the effort for kernel designing. The conventional multiple kernel learning methods learn a consensus kernel by linearly combining all candidate kernels, whereas ignoring the influence of the noises. To … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
3
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
6
1

Relationship

2
5

Authors

Journals

citations
Cited by 8 publications
(3 citation statements)
references
References 22 publications
0
3
0
Order By: Relevance
“…Note that the proposed algorithm integrates the MKL with local and global structure learning and the Hilbert space selfexpressiveness property in a unified optimization problem. A denoising MKL method was presented by Zhou et al [59]. It considers two kinds of noise: local noise, which appears in a small number of elements of the kernel matrix and is often induced by outliers or corrupted instances, and global noise, which appears in most of the elements of the kernel matrix and is often induced by inappropriate kernels.…”
Section: B Background Reviewmentioning
confidence: 99%
“…Note that the proposed algorithm integrates the MKL with local and global structure learning and the Hilbert space selfexpressiveness property in a unified optimization problem. A denoising MKL method was presented by Zhou et al [59]. It considers two kinds of noise: local noise, which appears in a small number of elements of the kernel matrix and is often induced by outliers or corrupted instances, and global noise, which appears in most of the elements of the kernel matrix and is often induced by inappropriate kernels.…”
Section: B Background Reviewmentioning
confidence: 99%
“…On the one hand, these datasets provide opportunities to reveal new insights into many biological problems, e.g., elucidating cell types, on the other hand, there are also computational challenges due to the amount of data. A straightforward approach to elucidate cell types in complex tissues is to partition the cells into some separated subgroups via clus-tering techniques [10]- [13], which can be regarded as an unsupervised classification problem [14]- [16]. Many previous clustering techniques can be used for this task, such as principal component analysis (PCA) [17], spectral clustering [18], and k-means [19].…”
Section: Introductionmentioning
confidence: 99%
“…The typical methods include K-means based [2,4,7,18,20,32,37,41], self-organizing map (SOM) [24], maximum margin clustering based [26,30,34], local learning-based [33], spectral clustering based [1,3,6,12,13,22,29] and subspace clustering based [8-10, 39, 40] algorithms. Compared with the single kernel counterpart, MKC should take special effort to handle the additional data problems such as noisy and incomplete kernels [2,15,19,21,28,36,38,40,41].…”
mentioning
confidence: 99%