Independent Component Analysis and Signal Separation: 7th International Conference, Ica 2007, London, Uk, September 9-12, 2007, Proceedings

ISBN-13: 9783540744931 / Angielski / Miękka / 2007 / 872 str.

Christopher C. James

Opis

This volume contains the papers presented at the 7th International Conference on Independent Component Analysis (ICA) and Source Separation held in L- don, 9 12 September 2007, at Queen Mary, University of London. Independent Component Analysis and Signal Separation is one of the most exciting current areas of research in statistical signal processing and unsup- vised machine learning. The area has received attention from several research communities including machine learning, neural networks, statistical signal p- cessing and Bayesian modeling. Independent Component Analysis and Signal Separation has applications at the intersection of many science and engineering disciplinesconcernedwithunderstandingandextractingusefulinformationfrom data as diverse as neuronal activity and brain images, bioinformatics, com- nications, the World Wide Web, audio, video, sensor signals, or time series. This year s event was organized by the EPSRC-funded UK ICA Research Network (www.icarn.org). There was also a minor change to the conference title this year with the exclusion of the word blind . The motivation for this was the increasing number of interesting submissions using non-blind or semi-blind techniques that did not really warrant this label. Evidence of the continued interest in the ?eld was demonstrated by the healthy number of submissions received, and of the 149 papers submitted just over two thirds were accepted."

Spis treści

Theory.- A Flexible Component Model for Precision ICA.- Blind Separation of Instantaneous Mixtures of Dependent Sources.- The Complex Version of the Minimum Support Criterion.- Optimal Joint Diagonalization of Complex Symmetric Third-Order Tensors. Application to Separation of Non Circular Signals.- Imposing Independence Constraints in the CP Model.- Blind Source Separation of a Class of Nonlinear Mixtures.- Independent Subspace Analysis Is Unique, Given Irreducibility.- Optimization on the Orthogonal Group for Independent Component Analysis.- Using State Space Differential Geometry for Nonlinear Blind Source Separation.- Copula Component Analysis.- On Separation of Signal Sources Using Kernel Estimates of Probability Densities.- Shifted Independent Component Analysis.- Modeling and Estimation of Dependent Subspaces with Non-radially Symmetric and Skewed Densities.- On the Relationships Between Power Iteration, Inverse Iteration and FastICA.- A Sufficient Condition for the Unique Solution of Non-Negative Tensor Factorization.- Colored Subspace Analysis.- Is the General Form of Renyi’s Entropy a Contrast for Source Separation?.- Algorithms.- A Variational Bayesian Algorithm for BSS Problem with Hidden Gauss-Markov Models for the Sources.- A New Source Extraction Algorithm for Cyclostationary Sources.- A Robust Complex FastICA Algorithm Using the Huber M-Estimator Cost Function.- Stable Higher-Order Recurrent Neural Network Structures for Nonlinear Blind Source Separation.- Hierarchical ALS Algorithms for Nonnegative Matrix and 3D Tensor Factorization.- Pivot Selection Strategies in Jacobi Joint Block-Diagonalization.- Speeding Up FastICA by Mixture Random Pruning.- An Algebraic Non Orthogonal Joint Block Diagonalization Algorithm for Blind Separation of Convolutive Mixtures of Sources.- Non Unitary Joint Block Diagonalization of Complex Matrices Using a Gradient Approach.- A Toolbox for Model-Free Analysis of fMRI Data.- An Eigenvector Algorithm with Reference Signals Using a Deflation Approach for Blind Deconvolution.- Robust Independent Component Analysis Using Quadratic Negentropy.- Underdetermined Source Separation Using Mixtures of Warped Laplacians.- Blind Separation of Cyclostationary Sources Using Joint Block Approximate Diagonalization.- Independent Process Analysis Without a Priori Dimensional Information.- An Evolutionary Approach for Blind Inversion of Wiener Systems.- A Complexity Constrained Nonnegative Matrix Factorization for Hyperspectral Unmixing.- Smooth Component Analysis as Ensemble Method for Prediction Improvement.- Speed and Accuracy Enhancement of Linear ICA Techniques Using Rational Nonlinear Functions.- Comparative Speed Analysis of FastICA.- Kernel-Based Nonlinear Independent Component Analysis.- Linear Prediction Based Blind Source Extraction Algorithms in Practical Applications.- Sparse Methods.- Blind Audio Source Separation Using Sparsity Based Criterion for Convolutive Mixture Case.- Maximization of Component Disjointness: A Criterion for Blind Source Separation.- Estimator for Number of Sources Using Minimum Description Length Criterion for Blind Sparse Source Mixtures.- Compressed Sensing and Source Separation.- Morphological Diversity and Sparsity in Blind Source Separation.- Identifiability Conditions and Subspace Clustering in Sparse BSS.- Two Improved Sparse Decomposition Methods for Blind Source Separation.- Probabilistic Geometric Approach to Blind Separation of Time-Varying Mixtures.- Infinite Sparse Factor Analysis and Infinite Independent Components Analysis.- Fast Sparse Representation Based on Smoothed ?0 Norm.- Estimating the Mixing Matrix in Sparse Component Analysis Based on Converting a Multiple Dominant to a Single Dominant Problem.- Dictionary Learning for L1-Exact Sparse Coding.- Supervised and Semi-supervised Separation of Sounds from Single-Channel Mixtures.- Image Compression by Redundancy Reduction.- Complex Nonconvex l p Norm Minimization for Underdetermined Source Separation.- Sparse Component Analysis in Presence of Noise Using an Iterative EM-MAP Algorithm.- Speech and Audio Applications.- Mutual Interdependence Analysis (MIA).- Modeling Perceptual Similarity of Audio Signals for Blind Source Separation Evaluation.- Beamforming Initialization and Data Prewhitening in Natural Gradient Convolutive Blind Source Separation of Speech Mixtures.- Blind Vector Deconvolution: Convolutive Mixture Models in Short-Time Fourier Transform Domain.- A Batch Algorithm for Blind Source Separation of Acoustic Signals Using ICA and Time-Frequency Masking.- The Role of High Frequencies in Convolutive Blind Source Separation of Speech Signals.- Signal Separation by Integrating Adaptive Beamforming with Blind Deconvolution.- Blind Signal Deconvolution as an Instantaneous Blind Separation of Statistically Dependent Sources.- Solving the Permutation Problem in Convolutive Blind Source Separation.- Discovering Convolutive Speech Phones Using Sparseness and Non-negativity.- Frequency-Domain Implementation of a Time-Domain Blind Separation Algorithm for Convolutive Mixtures of Sources.- Phase-Aware Non-negative Spectrogram Factorization.- Probabilistic Amplitude Demodulation.- First Stereo Audio Source Separation Evaluation Campaign: Data, Algorithms and Results.- ‘Shadow BSS’ for Blind Source Separation in Rapidly Time-Varying Acoustic Scenes.- Biomedical Applications.- Evaluation of Propofol Effects in Atrial Fibrillation Using Principal and Independent Component Analysis.- Space-Time ICA and EM Brain Signals.- Extraction of Gastric Electrical Response Activity from Magnetogastrographic Recordings by DCA.- ECG Compression by Efficient Coding.- Detection of Paroxysmal EEG Discharges Using Multitaper Blind Signal Source Separation.- Constrained ICA for the Analysis of High Stimulus Rate Auditory Evoked Potentials.- Gradient Convolution Kernel Compensation Applied to Surface Electromyograms.- Independent Component Analysis of Functional Magnetic Resonance Imaging Data Using Wavelet Dictionaries.- Multivariate Analysis of fMRI Group Data Using Independent Vector Analysis.- Extraction of Atrial Activity from the ECG by Spectrally Constrained ICA Based on Kurtosis Sign.- Blind Matrix Decomposition Techniques to Identify Marker Genes from Microarrays.- Perception of Transformation-Invariance in the Visual Pathway.- Subspaces of Spatially Varying Independent Components in fMRI.- Multi-modal ICA Exemplified on Simultaneously Measured MEG and EEG Data.- Miscellaneous.- Blind Signal Separation Methods for the Identification of Interstellar Carbonaceous Nanoparticles.- Specific Circumstances on the Ability of Linguistic Feature Extraction Based on Context Preprocessing by ICA.- Conjugate Gamma Markov Random Fields for Modelling Nonstationary Sources.- Blind Separation of Quantum States: Estimating Two Qubits from an Isotropic Heisenberg Spin Coupling Model.- An Application of ICA to BSS in a Container Gantry Crane Cabin’s Model.- Blind Separation of Non-stationary Images Using Markov Models.- Blind Instantaneous Noisy Mixture Separation with Best Interference-Plus-Noise Rejection.- Compact Representations of Market Securities Using Smooth Component Extraction.- Bayesian Estimation of Overcomplete Independent Feature Subspaces for Natural Images.- ICA-Based Image Analysis for Robot Vision.- Learning of Translation-Invariant Independent Components: Multivariate Anechoic Mixtures.- Channel Estimation for O-STBC MISO Systems Using Fourth-Order Cross-Cumulants.- System Identification in Structural Dynamics Using Blind Source Separation Techniques.- Image Similarity Based on Hierarchies of ICA Mixtures.- Text Clustering on Latent Thematic Spaces: Variants, Strengths and Weaknesses.- Top-Down Versus Bottom-Up Processing in the Human Brain: Distinct Directional Influences Revealed by Integrating SOBI and Granger Causality.- Noisy Independent Component Analysis as a Method of Rotating the Factor Scores.- Multilinear (Tensor) ICA and Dimensionality Reduction.- ICA in Boolean XOR Mixtures.- A Novel ICA-Based Image/Video Processing Method.- Keynote Talk.- Blind Audio Source Separation Based on Independent Component Analysis.