10,947 Hits in 6.2 sec

A class of neural networks for independent component analysis

J. Karhunen, E. Oja, L. Wang, R. Vigario, J. Joutsensalo
1997 IEEE Transactions on Neural Networks  
We consider learning algorithms for each layer, and modify our previous nonlinear PCA type algorithms so that their separation capabilities are greatly improved.  ...  The basic ICA network consists of whitening, separation, and basis vector estimation layers. It can be used for both blind source separation and estimation of the basis vectors of ICA.  ...  ACKNOWLEDGMENT The authors are grateful to the reviewers for their detailed and useful comments.  ... 
doi:10.1109/72.572090 pmid:18255654 fatcat:7ycqgkg5yvdhflruafe75wfaay

Automatic damage type classification and severity quantification using signal based and nonlinear model based damage sensitive features

Meriem Ghrib, Marc Rébillat, Guillaume Vermot des Roches, Nazih Mechbal
2018 Journal of Process Control  
Two types of features are used as inputs to the SVM algorithm: Signal Based Features (SBF) and Nonlinear Model Based Features (NMBF).  ...  [10] damage classification is performed using time-frequency representations and the Adaboost machine learning algorithm.  ... 
doi:10.1016/j.jprocont.2018.08.002 fatcat:ykst3cuwinanbooffot5suelxu

Cross Signal Identification For Psg Applications

Carmen Grigoraş, Victor Grigoraş, Daniela Boişteanu
2008 Zenodo  
An alternative representation of the respiratory events by means of Kohonen type neural network is discussed.  ...  Our computed analysis includes a learning phase based on cross signal PSG annotation.  ...  We tried sigmoid type nonlinearities for the representation of ECG with NLPCA in a previous work [2] , and also hysteresis type nonlinearity, which showed poor convergence to the eigen coordinate vectors  ... 
doi:10.5281/zenodo.1334641 fatcat:4l2aicpezzbifnyvz36abjtrlu

Harmonic Elimination of Inverters using Blind Signal Separation

B. Justus Rabi, R. Arumugam
2005 American Journal of Applied Sciences  
The harmonic separation process is implemented with a processor achieves low THD using Blind Signal separation. It is mostly used in medical instrumentation and medical applications like ECG, EEG.  ...  The main objective of this study is to eliminate harmonics and reduce the power loss in inverters.  ...  However this kind of representation often characterizes the fundamental properties of the data better than standard PCA. For example in blind signal separation of the original source signals.  ... 
doi:10.3844/ajassp.2005.1434.1437 fatcat:wocuz6r6efczfo67rla4vdpsca

Page 5268 of Psychological Abstracts Vol. 81, Issue 11 [page]

1994 Psychological Abstracts  
(Helsinki U of Technology, Lab of Computer & Information Science, Finland) Representation and separation of signals using nonlinear PCA type learning. Neural Networks, 1994, Vol 7(1), 113-127.  ...  —De- rives a new class of nonlinear principal component analysis (PCA) type learning algorithms by minimizing a general statisti- cal signal representation error.  ... 

Principal manifolds and Bayesian subspaces for visual recognition

B. Moghaddam
1999 Proceedings of the Seventh IEEE International Conference on Computer Vision  
We investigate the use of linear and nonlinear principal manifolds for learning low-dimensional representations for visual recognition.  ...  Abstract We i n vestigate the use of linear and nonlinear principal manifolds for learning lowdimensional representations for visual recognition.  ...  Since the Bayesian similarity method's learning stage requires two separate PCAs, its complexity is essentially twice that of PCA.  ... 
doi:10.1109/iccv.1999.790407 dblp:conf/iccv/Moghaddam99 fatcat:uew4mkxlvbeh7bdtjwu4fxg2b4

Automatic Damage Quantification Using Signal Based And Nonlinear Model Based Damage Sensitive Features

Meriem Ghrib, Marc Rébillat, Nazih Mechbal, Guillaume Vermot des Roches
2017 IFAC-PapersOnLine  
Two types of features are used as inputs to the SVM algorithm: Signal Based Features (SBF) and Nonlinear Model Based Features (NMBF).  ...  SBF are rooted in a direct use of response signals and do not consider any underlying model of the test structure.  ...  SVMs and PCA SVMs SVM learning technique is used for the classification step.  ... 
doi:10.1016/j.ifacol.2017.08.994 fatcat:acaz6htrajbcxahxhxoa3rwace

Neural Network Implementations for PCA and Its Extensions

Jialin Qiu, Hui Wang, Jiabin Lu, Biaobiao Zhang, K.-L. Du
2012 ISRN Artificial Intelligence  
These methods are useful in adaptive signal processing, blind signal separation (BSS), pattern recognition, and information compression.  ...  PCA is a statistical method that is directly related to EVD and SVD. Minor component analysis (MCA) is a variant of PCA, which is useful for solving total least squares (TLSs) problems.  ...  PCA is often used to select inputs, but it is not always useful, since the variance of a signal is not always related to the importance of the signal, for non-Gaussian signals.  ... 
doi:10.5402/2012/847305 fatcat:5v5l5v56ozg7lkxfktm5t7cgle

Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics

Christoph Wehmeyer, Frank Noé
2018 Journal of Chemical Physics  
Inspired by the success of deep learning techniques in the physical and chemical sciences, we apply a modification of an autoencoder type deep neural network to the task of dimension reduction of molecular  ...  the capabilities of linear dimension reduction techniques.  ...  European Commission (ERC StG 307494 "pcCell") and Deutsche Forschungsgemeinschaft (SFB 1114/A04).  ... 
doi:10.1063/1.5011399 pmid:29960344 fatcat:4lthaotfqfblfkh5uyaxmpgwza

Page 747 of Neural Computation Vol. 6, Issue 4 [page]

1994 Neural Computation  
Karhunen, J., and Joutsensalo, J. 1993. Representation and separation of signals using nonlinear PCA type learning. Neural Networks, in press. Oja, E., and Karhunen, J. 1985.  ...  Stability of Oja’s PCA Subspace Rule Acknowledgment The author is grateful to Jyrki Joutsensalo for providing the simulation example. References Hertz, J., Krogh, A., and Palmer, R. G. 1991.  ... 

Comparison of dimensionality reduction techniques for the fault diagnosis of mono block centrifugal pump using vibration signals

N.R. Sakthivel, Binoy B. Nair, M. Elangovan, V. Sugumaran, S. Saravanmurugan
2014 Engineering Science and Technology, an International Journal  
In this paper dimensionality reduction is performed using traditional dimensionality reduction techniques and nonlinear dimensionality reduction techniques.  ...  This is achieved by the extraction of features from the measured data and employing data mining approaches to explore the structural information hidden in the signals acquired.  ...  From Tables 1e4 among the nonlinear dimensionality reduction techniques and PCA, the PCA outperforms when using decision tree, Bayes Net, Naïve Bayes and kNN classifiers.  ... 
doi:10.1016/j.jestch.2014.02.005 fatcat:puldrr4n7vb63cbc2umaweoq64

Classification of the myoelectric signal using time-frequency based representations

K Englehart, B Hudgins, P.A Parker, M Stevenson
1999 Medical Engineering and Physics  
An accurate and computationally efficient means of classifying surface myoelectric signal patterns has been the subject of considerable research effort in recent years.  ...  Effective feature extraction is crucial to reliable classification and, in the quest to improve the accuracy of transient myoelectric signal pattern classification, an ensemble of time-frequency based  ...  Acknowledgements The authors acknowledge the assistance of the Natural Sciences and Engineering Research Council of Canada and the Whitaker Foundation.  ... 
doi:10.1016/s1350-4533(99)00066-1 pmid:10624739 fatcat:75ae6ty3rnhwdeknbjfiqwfvqe


1996 International Journal of Neural Systems  
Our learning scheme provides a way for balancing the cooperation and competition necessary for the self-organization process thus realizes the multiple causes model, which accounts for an observed data  ...  Comparing with previ ous probability theory based multiple causes models, our scheme is much easier to implement and quite reliable.  ...  Sometime they are treated as nonlinear extensions of some PCA learnings or shortly termed as nonlinear PCA 8?11 .  ... 
doi:10.1142/s0129065796000208 fatcat:mxrdvepc4zgxtehkbqsrhznxd4

Dimensionality Reduction and Anomaly Detection for CPPS Data using Autoencoder

Benedikt Eiteneuer, Nemanja Hranisavljevic, Oliver Niggemann
2019 2019 IEEE International Conference on Industrial Technology (ICIT)  
A closely related concern is dimensionality reduction (DR) which is: 1) often used as a preprocessing step in an AD solution, 2) a sort of AD, if a measure of observation conformity to the learned data  ...  Moreover, the approach outperforms state-of-the-art techniques, alongside a relatively simple and straightforward application.  ...  An autoencoder neural network or autoencoder [9] is a special type of deep feed-forward neural network, typically used for representation learning and dimensionality reduction.  ... 
doi:10.1109/icit.2019.8755116 dblp:conf/icit2/EiteneuerHN19 fatcat:pfmvi7am3fecpkxp2lqmignntm

Finding Clusters and Components by Unsupervised Learning [chapter]

Erkki Oja
2004 Lecture Notes in Computer Science  
In statistical PR, there are two classical categories for unsupervised learning methods and models: first, variations of Principal Component Analysis and Factor Analysis, and second, learning vector coding  ...  This approach is also reviewed, with examples such as linear and nonlinear independent component analysis and topological maps.  ...  of the separated signals y1, Separated signals 36]: 1.  ... 
doi:10.1007/978-3-540-27868-9_1 fatcat:e2yzfehty5bylca47aolqcnqnm
« Previous Showing results 1 — 15 out of 10,947 results