Double Robust Principal Component Analysis

Double Robust Principal Component Analysis
复制标题

双稳健主成分分析

DOI:
10.1016/j.neucom.2020.01.097
复制
发表时间:
2020-05
期刊:
影响因子:
6
通讯作者:
Chris Ding
Chris Ding
中科院分区:
计算机科学2区
文献类型:
--
作者:
Qianqian Wang;QuanXue Gao;Gan Sun;Chris Ding

文献摘要

参考文献

被引文献

相似文献

鲁棒主成分分析(RPCA)旨在从损坏的数据中恢复具有低秩结构的底层干净数据,是机器学习和数据挖掘中的强大工具。然而,在许多现实世界的应用中,测试阶段的新数据(即样本外)在训练过程中可能是看不见的,(1)RPCA作为一种转导方法自然无法处理样本外的问题,(2)将RPCA暴力地应用到这些应用中并没有明确考虑重建误差和低秩表示之间的关系。为了解决这些问题,在本文中,我们提出了一种双稳健主成分分析来处理样本外问题,称为 DRPCA。更具体地说,我们将重构误差集成到 RPCA 的标准函数中。然后,我们提出的模型可以受益于(1)主成分对异常值和缺失值的鲁棒性,(2)重建误差和低秩表示之间的桥梁,(3)通过线性变换从新数据中提取低秩干净数据。为此,在多个数据集上进行的广泛实验证明了其在多个聚类和低秩恢复任务中与最先进的模型相比的优越性。
Robust Principal Component Analysis (RPCA) aiming to recover underlying clean data with low-rank structure from the corrupted data, is a powerful tool in machine learning and data mining. However, in many real-world applications where new data (i.e., out-of-samples) in the testing phase can be unseen in the training procedure, (1) RPCA which is a transductive method can be naturally incapable of handing out-of-samples, and (2) violently applying RPCA into this applications does not explicitly consider the relationships between reconstruction error and low-rank representation. To tackle these problems, in this paper, we propose a Double Robust Principal Component Analysis to deal with the out-of-sample prob- lems, which is termed as DRPCA. More specifically, we integrate a reconstruction error into the criterion function of RPCA. Our proposed model can then benefit from (1) the robustness of principal components to outliers and missing values, (2) the bridge between reconstruction error and low-rank representation, (3) low-rank clean data extraction from new datum by a linear transform. To this end, extensive experiments on several datasets demonstrate its superiority, when comparing with the state-of-the-art models, in several clustering and low-rank recovery tasks.
DOI: 10.1609/aaai.v31i1.10798
发表时间: 2017-02
期刊: --
影响因子: --
作者:
Qianqian Wang;Quanxue Gao
通讯作者: Qianqian Wang;Quanxue Gao
DOI: 10.1007/bfb0015522
发表时间: 1996-04
期刊: --
影响因子: --
作者:
P. Belhumeur;J. Hespanha;D. Kriegman
通讯作者: P. Belhumeur;J. Hespanha;D. Kriegman
DOI: 10.1109/tpami.2012.274
发表时间: 2013-07
影响因子: 23.6
作者:
Zhenyue Zhang;Keke Zhao
通讯作者: Zhenyue Zhang;Keke Zhao
DOI: 10.1162/jocn.1991.3.1.71
发表时间: 1991-12-01
影响因子: 3.2
作者:
TURK, M;PENTLAND, A
通讯作者: PENTLAND, A
DOI: 10.1109/iccv.2005.167
发表时间: 2005-10
期刊: Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1
影响因子: --
作者:
Xiaofei He;Deng Cai;Shuicheng Yan;HongJiang Zhang
通讯作者: Xiaofei He;Deng Cai;Shuicheng Yan;HongJiang Zhang