A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2016; you can also visit the original URL.
The file type is application/pdf.
Filters
:{unav)
2012
Journal of VLSI Signal Processing Systems for Signal, Image and Video Technology
Preserved Fulltext
A novel algorithm based on Renyi's quadratic entropy is used to train, directly from a data set, linear or nonlinear mappers for entropy maximization or minimization. ...
This paper discusses a framework for learning based on information theoretic criteria. ...
Acknowledgments: This work was partially supported by a DARPA-Air Force grant F33615-97-1019 and NSF ECS-9900394. ...
doi:10.1023/a:1008143417156
fatcat:jk7422trdfaxzo4hvdicfwe7ei
An Information-Maximization Approach to Blind Separation and Blind Deconvolution
1995
Neural Computation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
We derive a new self-organizing learning algorithm that maximizes the information transferred in a network of nonlinear units. ...
Finally, we derive dependencies of information transfer on time delays. We suggest that information maximization provides a unifying framework for problems in "blind" signal processing. ...
Many helpful observations also came from Paul Viola, Barak Pearlmutter, Kenji Doya, Misha Tsodyks, Alexandre Pouget, Peter Dayan, Olivier Coenen, and Iris Ginzburg. ...
doi:10.1162/neco.1995.7.6.1129
pmid:7584893
fatcat:26psxygznbfhtbkha7q4fgjjee
An Introduction to Information Theoretic Learning, Part II: Applications
2016
Journal of Communication and Information Systems
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
This is the second part of the introductory tutorial about information theoretic learning, which, after the theoretical foundations presented in Part I, now discusses the concepts of correntropy, a new ...
similarity measure derived from the quadratic entropy, and presents example problems where the ITL framework can be successfully applied: dynamic modelling, equalization, independent component analysis ...
ACKNOWLEDGMENT The authors thank FAPESP (Grant 2013/14185-2), CAPES and CNPq for the financial support. ...
doi:10.14209/jcis.2016.7
fatcat:fjsomfgggfglvcpkklr4jvmrne
Entropy minimization for supervised digital communications channel equalization
2002
IEEE Transactions on Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Moreover, for a linear equalizer, an orthogonality condition for the minimum entropy solution that leads to an alternative fixed-point iterative minimization method is derived. ...
On the other hand, for nonlinear channels and using a multilayer perceptron (MLP) as the equalizer, differences between both criteria appear. ...
ACKNOWLEDGMENT The authors would like to thank the referees for providing us with valuable comments and insightful suggestions that have greatly improved this paper. ...
doi:10.1109/78.995074
fatcat:lmqwnkqxzzawlj6pnk6zgemige
An analysis of entropy estimators for blind source separation
2006
Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2011; you can also visit the original URL.
The file type is application/pdf.
Three reasons are given why Renyi's entropy estimators for Information-Theoretic Learning (ITL), on which the proposed method is based, is to be preferred over Shannon's entropy estimators for ITL. ...
An extensive analysis of a non-parametric, information-theoretic method for instantaneous blind source separation (BSS) is presented. ...
These three criteria will be referred to as the (modified) Minimum Renyi's Mutual Information (MRMI), Minimum Shannon Mutual Information (MSMI), and (modified) MRMI-SIG criteria, respectively. ...
doi:10.1016/j.sigpro.2005.04.015
fatcat:zuybddsqsbfczp24m5ppdbfj4y
Unsupervised Learning based Modified C- ICA for Audio Source Separation in Blind Scenario
2016
International Journal of Information Technology and Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
This work proposed divergence algorithm designed for faster convergence speed along with good quality of separation. ...
The main challenge in BASS is quality of separation and separation speed and the convergence speed gets compromised when separation techniques focused on quality of separation. ...
Estimation of difference between joint entropy and marginal entropy of different information sources leads to ICA implementation using minimum mutual information (MMI). ...
doi:10.5815/ijitcs.2016.03.02
fatcat:346wb4p2ebb6poyyipi5g4xgiu
Blind source separation of convolutive mixtures
2006
Independent Component Analyses, Wavelets, Unsupervised Smart Sensors, and Neural Networks IV
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2011; you can also visit the original URL.
The file type is application/pdf.
BSS can be regarded as an intelligent version of ABF in the sense that it can adapt without any information on the array manifold or the target direction, and sources can be simultaneously active in BSS ...
This paper introduces the blind source separation (BSS) of convolutive mixtures of acoustic signals, especially speech. ...
We search for the separation matrix W(ω) that minimizes the mutual information, maximize the nongaussianity, or maximize the likelihood of the output. ...
doi:10.1117/12.674413
fatcat:rjp6ttufbnf37fajrpt2msqaxy
Adaptive blind signal processing-neural network approaches
1998
Proceedings of the IEEE
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2005; you can also visit the original URL.
The file type is application/pdf.
Learning algorithms and underlying basic mathematical ideas are presented for the problem of adaptive blind signal processing, especially instantaneous blind separation and multichannel blind deconvolution ...
We discuss recent developments of adaptive learning algorithms based on the natural gradient approach and their properties concerning convergence, stability, and efficiency. ...
Back, and Dr. N. Murata for their fruitful collaboration and helpful discussions. ...
doi:10.1109/5.720251
fatcat:jg337aeuxnd3rec634qd3qjfde
Convergence properties and data efficiency of the minimum error entropy criterion in adaline training
2003
IEEE Transactions on Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2004; you can also visit the original URL.
The file type is application/pdf.
Recently, we have proposed the minimum error entropy (MEE) criterion as an information theoretic alternative to the widely used mean square error criterion in supervised adaptive system training. ...
Mathematical investigation of the proposed entropy estimator revealed interesting insights about the process of information theoretical learning. ...
Other successful applications of the proposed nonparametric entropy estimator and MEE include maximally informative subspace projections, blind source separation [9] - [11] , and blind deconvolution ...
doi:10.1109/tsp.2003.812843
fatcat:bzwfdf2i3zdsvgxblyrfuehfnq
Self-Adaptive Blind Source Separation Based on Activation Functions Adaptation
2004
IEEE Transactions on Neural Networks
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The learning algorithm for the activation function adaptation is consistent with the one for training the demixing model. ...
paper to develop a general framework of blind separation from a practical point of view with special emphasis on the activation function adaptation. ...
Various approaches, such as entropy maximization and minimization of mutual information, lead to the cost function (4) where is determined adaptively during training. ...
doi:10.1109/tnn.2004.824420
pmid:15384517
fatcat:kwskyv54kvdzzil7mejvlfflwa
Comparison of maximum entropy and minimal mutual information in a nonlinear setting
2002
Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2007; you can also visit the original URL.
The file type is application/pdf.
In blind source separation (BSS), two di erent separation techniques are mainly used: minimal mutual information (MMI), where minimization of the mutual output information yields an independent random ...
vector, and maximum entropy (ME), where the output entropy is maximized. ...
Acknowledgements We thank the referees for their helpful comments during the preparation of this paper. ...
doi:10.1016/s0165-1684(02)00200-1
fatcat:eoufootd7fhc7hmgruyosbj3mm
Blind Identification and Separation of Noisy Source Signals : Neural Network Approaches
1998
Systems, Control and Information
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Sympo- In this paper, we have presented neural net-IVOLTItl-9Z pp. 731-734 (1997) works On-line adaptive algorithms models and a family of associated adaptive in non stationary en- learning algorithms ...
Emphasis was given to the neural network or adaptive multichannel filtering models and associated on-line nonlinear adaptive learn- ing algorithms which have some biological resem- blance or plausibility ...
, Control and Information Engineers Institute ofSystems, Control and Information Engineers ...
doi:10.11509/isciesci.42.2_63
fatcat:6xkansmiungsbngdudq2pxmwpy
Fast and robust fixed-point algorithms for independent component analysis
1999
IEEE Transactions on Neural Networks
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2012; you can also visit the original URL.
The file type is application/pdf.
and/or of minimum variance. ...
These contrast functions enable both the estimation of the whole decomposition by minimizing mutual information, and estimation of individual independent components as projection pursuit directions. ...
The advantage of neural on-line learning rules is that the inputs can be used in the algorithm at once, thus enabling faster adaptation in a nonstationary environment. ...
doi:10.1109/72.761722
pmid:18252563
fatcat:5jngho43xfhs3jcs4st7cogx7q
Independent Component Analysis and Blind Signal Separation: Theory, Algorithms and Applications
2012
Learning and Nonlinear Models
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
An overview on the main statistical principles that guide the search for the independent components is formulated, methods for blind signal separation that require both high-order and second-order statistics ...
Some of the most successful algorithms for both ICA and BSS are derived. ...
Acknowledgements The authors are thankful for the support provided by CNPq and FAPERJ (Brazil), and for the Brazilian Navy Research Institute (IPqM) for providing the data set used in this work. ...
doi:10.21528/lnlm-vol10-no1-art4
fatcat:fewa5i5dozbilbr3euyntf4kvu
Electrical Power System Harmonic Analysis Technology Based on Fast ICA BSS Algorithm
2013
Advances in Information Sciences and Service Sciences
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Through comparing the separation performance by using the different step size, the adaptive changing step size natural gradient blind separation algorithm for the electrical power system harmonic signal ...
The experiment result proved that the electrical power system harmonic signal separation based on the gradient algorithm is accurate. ...
Blind signal separation algorithm mainly includes: the information maximization (Informax) algorithm [4] , the natural gradient algorithm, the Equivariant Adaptive Blind Separation (EASI) algorithm [ ...
doi:10.4156/aiss.vol5.issue7.6
fatcat:kewkmkslkjbmpnqstd3q5fd6gu
« Previous
Showing results 1 — 15 out of 1,795 results