474,974 Hits in 6.3 sec

A parallel independent component analysis algorithm

Hongtao Du, Hairong Qi, Xiaoling Wang
2006 12th International Conference on Parallel and Distributed Systems - (ICPADS'06)  
Independent Component Analysis (ICA), orienting as an efficient approach to the blind source separation (BSS) problem, searches for a linear or nonlinear transformation that minimizes the statistical dependence  ...  In this paper, a SPMD-structured parallel ICA (pICA) algorithm is presented. pICA is developed based on the FastICA approach and conducted in three stages: the estimation of weight matrix in which sub-processes  ...  Introduction Independent Component Analysis (ICA) is a method that searches for a linear or nonlinear non-orthogonal coordinate system in any multivariate data, in which the directions of the axes are  ... 
doi:10.1109/icpads.2006.17 dblp:conf/icpads/DuQW06 fatcat:7iw6geolerd3tnb6jpyeattxru

Parallel ICA methods for EEG neuroimaging

D.B. Keith, C.C. Hoge, R.M. Frank, A.D. Malony
2006 Proceedings 20th IEEE International Parallel & Distributed Processing Symposium  
HiPerSAT, a C++ library and tools, processes EEG data sets with ICA (Independent Component Analysis) methods.  ...  ICA is a class of methods for analyzing a large set of data samples and extracting independent components that explain the observed data.  ...  This research was partially funded by a grant from the National Science Foundation, Major Research Instrumentation program, and a contract from the Department of Defense, Telemedicine Advanced Technology  ... 
doi:10.1109/ipdps.2006.1639299 dblp:conf/ipps/KeithHFM06 fatcat:eoylbybfybbutlecuzekyfqrhi

Convergence analysis for a class of source separation methods

Alper T. Erdogan
2011 2011 XXIII International Symposium on Information, Communication and Automation Technologies  
This article reviews some convergence analysis results for parallel BSS algorithms that are used for the simultaneous extraction of sources.  ...  The emphasis will be on two major BSS schemes, namely, Indepenent Component Analysis and Bounded Component Analysis.  ...  Among them, Independent Component Analysis is the well established and most popular scheme whereas the other approach, namely the Bounded Component Analysis, is a recently introduced and a promising technique  ... 
doi:10.1109/icat.2011.6102083 dblp:conf/icatech/Erdogan11 fatcat:r4x6i4nze5hijfsa4et6injo4y

Three-way parallel independent component analysis for imaging genetics using multi-objective optimization

Alvaro Ulloa, Jingyu Liu, Victor Vergara, Jiayu Chen, Vince Calhoun, Marios Pattichis
2014 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society  
This paper describes a new algorithm, three-way parallel independent component analysis (3pICA), for jointly identifying genomic loci associated with brain function and structure.  ...  Methods based on independent component analysis have proven to be effective in jointly analyzing multiple modalities, including brain imaging and genetic data.  ...  Independent Component Analysis Matrix decomposition X = AS S: row sources,the weighted pattern of variables. A: how each source is represented across subjects.  ... 
doi:10.1109/embc.2014.6945153 pmid:25571521 pmcid:PMC6353612 fatcat:lbfp6w3ko5g4vpfkcudt4yal4y

Parallel group independent component analysis for massive fMRI data sets

Shaojie Chen, Lei Huang, Huitong Qiu, Mary Beth Nebel, Stewart H. Mostofsky, James J. Pekar, Martin A. Lindquist, Ani Eloyan, Brian S. Caffo, Satoru Hayasaka
2017 PLoS ONE  
To address this problem, we propose a two-stage likelihood-based algorithm for performing group ICA, which we denote Parallel Group Independent Component Analysis (PGICA).  ...  Formal analysis: SC LH HQ AE. Funding acquisition: BC. Investigation: MB SM JP AE. Parallel group independent component analysis for massive fMRI data sets PLOS ONE |  ...  The proposed Parallel Group Independent Component Analysis (PGICA) is different from fastICA and JADE in that the algorithm is likelihood-based and uses maximum likelihood estimation (MLE) for parameter  ... 
doi:10.1371/journal.pone.0173496 pmid:28278208 pmcid:PMC5344430 fatcat:yd572ihc4zcajp6pdxy4rlnsra

On blind separation of complex-valued sources by extended Hebbian learning

S. Fiori
2001 IEEE Signal Processing Letters  
The aim of this letter is to present a nonlinear extension to Sanger's generalized Hebbian learning algorithm for complex-valued data neural processing, which allows for separating mixed independent circular  ...  This principle was recently extended by the present author [8] to a three-layer network for performing information-theoretic-based independent component analysis.  ...  In this letter, we formally derive a new learning algorithm as a nonlinear complex extension of generalized Hebbian algorithm [10] for a linear feedforward network, the extended Hebbian learning algorithm  ... 
doi:10.1109/97.935735 fatcat:x2auyjxnufblbbcrbohd5t472q

Gaudi components for concurrency: Concurrency for existing and future experiments

M Clemencic, D Funke, B Hegner, P Mato, D Piparo, I Shapoval
2015 Journal of Physics, Conference Series  
In this paper, we present components introduced to express and track dependencies of algorithms to deduce a precedence-constrained directed acyclic graph, which serves as basis for our algorithmically  ...  sequence optimization by graph analysis.  ...  This entails handling several events concurrently within one process -inter-event parallelism -as well as simultaneously executing independent algorithms for one event -intra-event parallelism.  ... 
doi:10.1088/1742-6596/608/1/012021 fatcat:3drot2bouna3zh3khpaqflebdu

Accelerator of Stacked Convolutional Independent Subspace Analysis for Deep Learning-Based Action Recognition

Lu He, Yan Luo, Yu Cao
2014 2014 IEEE 22nd Annual International Symposium on Field-Programmable Custom Computing Machines  
We design parallel pipelines, data parallelisms and look-up table to speed up the algorithm.  ...  Recent advances in deep learning combined with stacked convolutional Independent Subspace Analysis (ISA) has achieved a better performance superior to all previously published results on several public  ...  In this paper, we focus on accelerating the stacked convolutional Independent Subspace Analysis (ISA) algorithm [1] .  ... 
doi:10.1109/fccm.2014.37 dblp:conf/fccm/HeLC14 fatcat:o44rb5mswvbkbj4hmniyoplmm4

A hierarchical approach to workload characterization for parallel systems [chapter]

M. Calzarossa, G. Haring, G. Kotsis, A. Merlo, D. Tessera
1995 Lecture Notes in Computer Science  
We propose a hierarchical approach to the systematic characterization of the workload of a parallel system, to be kept as modular and exible as possible.  ...  For each o f t h e s e l a yers di erent c haracteristics representing functional, sequential, parallel, and quantitative descriptions have been identi ed.  ...  By looking at the internal behavior of the algorithms A 1 and A 2 we nd out that they are completely independent o f e a c h other.  ... 
doi:10.1007/bfb0046616 fatcat:nmm4mvkj25dezl25ht7du5zpwe

Performance comparison of parallel fastICA algorithm in the PLGrid structures

Anna Gajos-Balinska, Grzegorz M Wojcik, Przemyslaw Stpiczynski, T. Kwater, B. Twarog, Z. Gomolka, J. Bartman
2018 ITM Web of Conferences  
One of the most commonly used methods is ICA (independent component analysis) [1] [2] [3] . However, algorithms of this type are computationally expensive.  ...  This paper presents a parallel implementation of the fastICA algorithm using the available libraries and extensions of the Intel processors (such as BLAS, MKL, Cilk Plus) and compares the execution time  ...  Therefore the task of the algorithm is to separate the independent components [12] .  ... 
doi:10.1051/itmconf/20182100026 fatcat:cuw2zs22tbadnigcx7nyl6lgky

Scalable Independent Multi-level Distribution in Multimedia Content Analysis [chapter]

Viktor S. Wold Eide, Frank Eliassen, Ole-Christoffer Granmo, Olav Lysne
2002 Lecture Notes in Computer Science  
In this paper we propose a component-based framework where each logical level can be parallelized and distributed independently.  ...  Due to the limited processing resources available on a typical host, monolithic multimedia content analysis applications are often restricted to simple content analysis tasks, covering a small number of  ...  This is the case when the content analysis task can be split into independent content analysis sub tasks.  ... 
doi:10.1007/3-540-36166-9_4 fatcat:7cajb57n45gs7ifrao6zosk2wy

Separation of Linearly Mixed Speech Signals using DWT based ICA

Daljeet Singh, Jaspinder Singh
2013 International Journal of Computer Applications  
Separating various linearly mixed speech signals is often modelled by famous cocktail party problem and can be achieved by a technique known as Independent Component Analysis (ICA).  ...  Comparison of existing ICA technique with the one proposed is done based on experimental results which shows that the proposed algorithm over performs basic ICA. 29 Results of extracted independent signals  ...  In deflationary approach, independent components are estimated one by one using the orthogonalization While in symmetric approach, independent components are estimated in parallel.  ... 
doi:10.5120/12256-8348 fatcat:bdzsxgzl3zcklhpgbyotydmipe

The ICA of Gravity Earth Tide Based on Genetic Algorithm and Extraction of Seismic Precursor Information

Xiao-Liang TANG, Hai-Yan QUAN
2017 DEStech Transactions on Computer Science and Engineering  
In the analysis of the Gravity Earth Tide signal, using a three-dimensional orthogonal decomposition model to decompose the harmonic information of the tidal signal into 3 orthogonal components.  ...  And this paper uses the traditional ICA and real coded genetic algorithm (GA) combined method, which has been proved is feasible and effective by the experiments .By the way we used the combination of  ...  Decomposition Model of the Gravity Earth Tidal Independent Component Analysis (ICA) Based on Real Code Genetic Algorithm ICA is a method for processing the blind signal which can separate an unknown  ... 
doi:10.12783/dtcse/aice-ncs2016/5650 fatcat:hlz2445gy5a7dlrwr25lqqsfxm

Parallel Model Checking and the FMICS-jETI Platform

Jiri Barnat, Lubos Brim, Martin Leucker
2007 12th IEEE International Conference on Engineering Complex Computer Systems (ICECCS 2007)  
In this paper we summarize parallel algorithms for enumerative model checking of properties formulated in linear time temporal logic (LTL) as well as a fragment of the µcalculus which naturally subsumes  ...  Therefore, it is desirable to provide parallel model checking applications as services for direct use and simple integration to customized modelling, analysis, and verification tools. jETI is a framework  ...  out a customized analysis algorithm.  ... 
doi:10.1109/iceccs.2007.34 dblp:conf/iceccs/BarnatBL07 fatcat:m5e7uf4izndw7krijcytfeuht4

Cost Estimation of Parallel Constrained Producer-Consumer Algorithms

Tariq Kamal, Keith R. Bisset, Ali R. Butt, Madhav Marathe
2015 2015 23rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing  
Cost estimation is crucial in the performance modeling of parallel algorithms and allocation of computational resources on distributed systems.  ...  This paper presents a novel methodology for estimating the cost of constrained producerconsumer (CPC) algorithms.  ...  Fig. 3 . 3 The general CPC parallel-algorithm, where c represents a task.  ... 
doi:10.1109/pdp.2015.115 dblp:conf/pdp/KamalBBM15 fatcat:msv7o6jx3bdn5hfaoavm3vxte4
« Previous Showing results 1 — 15 out of 474,974 results