Nonlinear dimensionality reduction using a temporal coherence principle

Huang, Yaping; Zhao, Jiali; Liu, Yunhui; Luo, Siwei; Zou, Qi; Mao, Tian

doi:10.1016/j.ins.2011.04.001

Cited by 25 publications

(10 citation statements)

References 46 publications

(60 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we evaluate the performance of our method ASFDA in comparison with other classical dimensionality reduction methods including LDA [2], DLPP [10], MFA [12], MMC [16], SSFACM [29], SSFASP [30], SFDA [31]and SSFA [32] on several publicly available databases. In order to make the comparison fair, we first apply PCA as preprocessing step to keep 98% energy.…”

Section: Experimental Results and Analysismentioning

confidence: 99%

“…SFA also has found many applications in the field of computational neuroscience [24,25] and time series analysis [26,27]. Several researchers introduce the slowness principle to the applications of pattern recognition [28][29][30][31][32]. Zhang and Tao [28] have successfully introduced the SFA framework to deal with the problem of human action recognition.…”

Section: Introductionmentioning

confidence: 99%

“…In [30], the authors propose another variant of Supervised Slow Feature that seeks the Shortest Path of each class samples (SSFASP) to construct time series for dimensionality reduction. Huang et al [31,32] utilize KNN criterion to construct time series and introduce Supervised Slow Feature Analysis (SSFA) for nonlinear dimensionality reduction. In order to get discriminant slow feature, they also propose slow feature discriminant analysis (SFDA) [31], which minimizes within-class temporal variation and maximizes between-class temporal variation simultaneously.…”

Section: Introductionmentioning

confidence: 99%

“…It is crucial to characterize suband multi-manifold information respectively. In the literatures [3,4,31,32], there are two common strategies for selecting withinclass time series: the k-nearest-neighborhood (k-NN), ε-neighborhood (ε-N). But they have distinct disadvantages.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Feature extraction using adaptive slow feature discriminant analysis

Liu

Wang

et al. 2015

Neurocomputing

View full text Add to dashboard Cite

Section: Experimental Results and Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Feature extraction using adaptive slow feature discriminant analysis

Liu

Wang

et al. 2015

Neurocomputing

View full text Add to dashboard Cite

“…To deal with nonlinear structural data, another type of technology is kernel-based methods. Huang et al 31 proposed a nonlinear dimensionality reduction framework based on the temporal coherence principle. Kernel-based methods have successfully extracted nonlinear features by applying linear techniques in the implicit feature space.…”

Section: Introductionmentioning

confidence: 99%

Adaptive unsupervised slow feature analysis for feature extraction

Liu

Wang

2015

J. Electron. Imaging

View full text Add to dashboard Cite

Slow feature analysis (SFA) extracts slowly varying features out of the input data and has been successfully applied on pattern recognition. However, SFA heavily relies on the constructed time series when SFA is applied on databases that neither have obvious temporal structure nor have label information. Traditional SFA constructs time series based on k-nearest neighborhood (k -NN) criterion. Specifically, the time series set constructed by k -NN criterion is likely to include noisy time series or lose suitable time series because the parameter k is difficult to determine. To overcome these problems, a method called adaptive unsupervised slow feature analysis (AUSFA) is proposed. First, AUSFA designs an adaptive criterion to generate time series for characterizing submanifold. The constructed time series have two properties: (1) two points of time series lie on the same submanifold and (2) the submanifold of the time series is smooth. Second, AUSFA seeks projections that simultaneously minimize the slowness scatter and maximize the fastness scatter to extract slow discriminant features. Extensive experimental results on three benchmark face databases demonstrate the effectiveness of our proposed method. Adaptive unsupervised slow feature analysis for feature extraction Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 05/14/2015 Terms of Use: http://spiedl.org/terms Adaptive unsupervised slow feature analysis for feature extraction Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 05/14/2015 Terms of Use: http://spiedl.org/terms Sheng Wang received his BS degree in automation from Henan University, China, in 2008. He obtained his MS degree in control theory and control engineering from the same university. Currently, he is a PhD student at Nanjing University of Science and Technology. His research interests include image processing, pattern recognition, and machine learning. Adaptive unsupervised slow feature analysis for feature extraction Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 05/14/2015 Terms of Use: http://spiedl.org/terms

show abstract

Dynamic nonlinear batch process fault detection and identification based on two‐directional dynamic kernel slow feature analysis

et al. 2020

View full text Add to dashboard Cite

The batch process generally covers high nonlinearity and two-directional dynamics: time-wise dynamics, which correspond to inherently time-varying dynamics resulting from the slowly varying underlying driving forces within each batch duration; and batch-wise dynamics, which are associated with different operating modes among different batches. However, most existing dynamic nonlinear monitoring methods cannot extract the slowly varying underlying driving forces of the nonlinear batch process and rarely tackle the batch-wise dynamic characteristics among batch runs. In order to address these issues, a new monitoring scheme based on two-directional dynamic kernel slow feature analysis (TDKSFA) is developed by combining kernel SFA with a global modelling strategy. In the TDKSFA method, kernel SFA is integrated with the ARMAX time series model to mine the nonlinear and timewise dynamic properties within a batch run due to its capability of extracting the slowly varying underlying driving forces. Furthermore, the global modelling strategy is presented to handle the batch-wise dynamics among batches by calculating the total average kernel matrix of all training batches. After the slow features are extracted, Hotelling's T 2 and SPE statistics are built to detect faults. To solve the issue of fault variable nonlinear identification, a novel nonlinear contribution plot inspired by the pseudo-sample variable projection trajectories in the TDKSFA model is further proposed to identify fault variables. Finally, the feasibility and effectiveness of the TDKSFA-based fault diagnosis strategy is demonstrated through a numerical system and the penicillin fermentation process. K E Y W O R D S batch process, fault diagnosis, slow feature analysis, two-directional dynamics 1 | INTRODUCTION Batch processes are becoming increasingly important in the pharmaceutical, semiconductor, polymer, and biochemical industries to manufacture high value-added products. However, it is difficult to establish a batch process monitoring model due to the high nonlinearity in time-wise and the time-varying dynamics both time-wise

show abstract

Nonlinear dimensionality reduction using a temporal coherence principle

Cited by 25 publications

References 46 publications

Feature extraction using adaptive slow feature discriminant analysis

Feature extraction using adaptive slow feature discriminant analysis

Adaptive unsupervised slow feature analysis for feature extraction

Dynamic nonlinear batch process fault detection and identification based on two‐directional dynamic kernel slow feature analysis

Contact Info

Product

Resources

About