Adaptive robust principal component analysis
- PMID: 31401529
- DOI: 10.1016/j.neunet.2019.07.015
Adaptive robust principal component analysis
Abstract
Robust Principal Component Analysis (RPCA) is a powerful tool in machine learning and data mining problems. However, in many real-world applications, RPCA is unable to well encode the intrinsic geometric structure of data, thereby failing to obtain the lowest rank representation from the corrupted data. To cope with this problem, most existing methods impose the smooth manifold, which is artificially constructed by the original data. This reduces the flexibility of algorithms. Moreover, the graph, which is artificially constructed by the corrupted data, is inexact and does not characterize the true intrinsic structure of real data. To tackle this problem, we propose an adaptive RPCA (ARPCA) to recover the clean data from the high-dimensional corrupted data. Our proposed model is advantageous due to: (1) The graph is adaptively constructed upon the clean data such that the system is more flexible. (2) Our model simultaneously learns both clean data and similarity matrix that determines the construction of graph. (3) The clean data has the lowest-rank structure that enforces to correct the corruptions. Extensive experiments on several datasets illustrate the effectiveness of our model for clustering and low-rank recovery tasks.
Keywords: Adaptively; Flexibility; RPCA.
Copyright © 2019 Elsevier Ltd. All rights reserved.
Similar articles
-
Robust Graph Learning From Noisy Data.IEEE Trans Cybern. 2020 May;50(5):1833-1843. doi: 10.1109/TCYB.2018.2887094. Epub 2019 Jan 8. IEEE Trans Cybern. 2020. PMID: 30629527
-
Inductive robust principal component analysis.IEEE Trans Image Process. 2012 Aug;21(8):3794-800. doi: 10.1109/TIP.2012.2192742. Epub 2012 Apr 3. IEEE Trans Image Process. 2012. PMID: 22481823
-
Low-rank representation with adaptive graph regularization.Neural Netw. 2018 Dec;108:83-96. doi: 10.1016/j.neunet.2018.08.007. Epub 2018 Aug 14. Neural Netw. 2018. PMID: 30173056
-
Graph Convolution RPCA With Adaptive Graph.IEEE Trans Image Process. 2022;31:6062-6071. doi: 10.1109/TIP.2022.3195669. Epub 2022 Sep 22. IEEE Trans Image Process. 2022. PMID: 35939456
-
Variable Selection for Time-to-Event Data.Methods Mol Biol. 2021;2194:61-76. doi: 10.1007/978-1-0716-0849-4_5. Methods Mol Biol. 2021. PMID: 32926362 Review.
Cited by
-
Special transcriptome landscape and molecular prognostic signature of non-smoking head and neck cancer patients.Funct Integr Genomics. 2023 Mar 8;23(2):79. doi: 10.1007/s10142-023-01002-6. Funct Integr Genomics. 2023. Retraction in: Funct Integr Genomics. 2024 May 13;24(3):93. doi: 10.1007/s10142-024-01367-2. PMID: 36882550 Retracted.
-
Robust Principal Component Thermography for Defect Detection in Composites.Sensors (Basel). 2021 Apr 10;21(8):2682. doi: 10.3390/s21082682. Sensors (Basel). 2021. PMID: 33920261 Free PMC article.
MeSH terms
LinkOut - more resources
Full Text Sources