EVD dualdating based online subspace learning (Q1718370): Difference between revisions
From MaRDI portal
Created claim: Wikidata QID (P12): Q59067114, #quickstatements; #temporary_batch_1711439739529 |
ReferenceBot (talk | contribs) Changed an Item |
||
Property / cites work | |||
Property / cites work: Principal component analysis. / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3998716 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4856771 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4841334 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Sequential Karhunen-Loeve basis extraction and its application to images / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: On Updating Problems in Latent Semantic Indexing / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Fast low-rank modifications of the thin singular value decomposition / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: An elementary proof of a theorem of Johnson and Lindenstrauss / rank | |||
Normal rank |
Latest revision as of 03:25, 18 July 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | EVD dualdating based online subspace learning |
scientific article |
Statements
EVD dualdating based online subspace learning (English)
0 references
8 February 2019
0 references
Summary: Conventional incremental PCA methods usually only discuss the situation of adding samples. In this paper, we consider two different cases: deleting samples and simultaneously adding and deleting samples. To avoid the NP-hard problem of downdating SVD without right singular vectors and specific position information, we choose to use EVD instead of SVD, which is used by most IPCA methods. First, we propose an EVD updating and downdating algorithm, called EVD dualdating, which permits simultaneous arbitrary adding and deleting operation, via transforming the EVD of the covariance matrix into a SVD updating problem plus an EVD of a small autocorrelation matrix. A comprehensive analysis is delivered to express the essence, expansibility, and computation complexity of EVD dualdating. A mathematical theorem proves that if the whole data matrix satisfies the low-rank-plus-shift structure, EVD dualdating is an optimal rank-\(k\) estimator under the sequential environment. A selection method based on eigenvalues is presented to determine the optimal rank \(k\) of the subspace. Then, we propose three incremental/decremental PCA methods: EVDD-IPCA, EVDD-DPCA, and EVDD-IDPCA, which are adaptive to the varying mean. Finally, plenty of comparative experiments demonstrate that EVDD-based methods outperform conventional incremental/decremental PCA methods in both efficiency and accuracy.
0 references