By Topic

2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing

Date 6-8 Sept. 2006

Filter Results

Displaying Results 1 - 25 of 90
  • Covers

    Publication Year: 2006, Page(s): C1
    Request permission for commercial reuse | PDF file iconPDF (65 KB)
    Freely Available from IEEE
  • Preface

    Publication Year: 2006, Page(s): nil1
    Request permission for commercial reuse | PDF file iconPDF (65 KB)
    Freely Available from IEEE
  • MLSP 2006 Technical Committee

    Publication Year: 2006, Page(s): nil2
    Request permission for commercial reuse | PDF file iconPDF (74 KB)
    Freely Available from IEEE
  • Table of contents

    Publication Year: 2006
    Request permission for commercial reuse | PDF file iconPDF (4152 KB)
    Freely Available from IEEE
  • Signal Detection, Pattern Recognition and Classification

    Publication Year: 2006, Page(s): 1
    Request permission for commercial reuse | PDF file iconPDF (182 KB)
    Freely Available from IEEE
  • Building Efficient Radial Basis Function Kernel Classifiers using Iterative Methods

    Publication Year: 2006, Page(s):3 - 8
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (7118 KB) | HTML iconHTML

    Training algorithms for radial basis function Kernel classifiers (RBFKCs), such as the canonical support vector machine (SVM), often produce computationally burdensome classifiers when large training data sets are used. Additionally, this complexity is not directly controllable by the developer. A least-squares variant of the SVM is used as a starting point for a proposed algorithm called the incr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The Correntropy Mace Filter for Image Recognition

    Publication Year: 2006, Page(s):9 - 14
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6192 KB) | HTML iconHTML

    The minimum average correlation energy (MACE) filter is a well known correlation filter for pattern recognition. This paper proposes a nonlinear extension to the MACE filter using the recently introduced correntropy function in feature space. Correntropy is a positive definite function that generalizes the concept of correlation by utilizing higher order moment information of signal structure. Sin... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Constraint-Based, Transductive Learning for Distributed Ensemble Classification

    Publication Year: 2006, Page(s):15 - 20
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6700 KB) | HTML iconHTML

    We consider ensemble classification when there is no common labeled data for designing the function which aggregates classifier decisions. In recent work, we dubbed this problem distributed ensemble classification, addressing e.g. when local classifiers are trained on different (e.g. proprietary, legacy) databases or operate on different sensing modalities. Typically, fixed (untrained) rules of cl... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Use Your Powers Wisely: Resource Allocation in Parallel Channels

    Publication Year: 2006, Page(s):21 - 26
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5894 KB) | HTML iconHTML

    This study evaluates various resource allocation strategies for simultaneous estimation of two independent signals from noisy observations. We focus on strategies that make use of the underlying dynamics of each signal, exploiting the difference in estimation uncertainty between them. This evaluation is done empirically, by exploring the parameter space through computer simulations. Two cases are ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Gibbsboost: a Boosting Algorithm using a Sequential Monte Carlo Approach

    Publication Year: 2006, Page(s):259 - 264
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6366 KB) | HTML iconHTML

    This study proposes a novel boosting algorithm, GibbsBoost. A Gibbs distribution of a weaklearner sequence with a specific loss (energy) function is used in this algorithm as the posterior distribution in Bayesian learning. Weaklearner sequence samples are recursively drawn from the distribution via sequential Monte Carlo. The predictions are derived from a combination of the weaklearner sequence ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic Image Classification by a Granular Computing Approach

    Publication Year: 2006, Page(s):33 - 38
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6602 KB) | HTML iconHTML

    In this paper we propose an image classification system able to solve automatically a large set of problem instances by a granular computing approach. By means of a watershed segmentation algorithm, each image is decomposed into a set of segments (information granules), characterized by suited color, texture and shape features (segment signature). Successively, images are represented by a symbolic... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Blind Signal Separation I

    Publication Year: 2006, Page(s): 39
    Request permission for commercial reuse | PDF file iconPDF (161 KB)
    Freely Available from IEEE
  • A Least Absolute Bound Approach to ICA - Application to the MLSP 2006 Competition

    Publication Year: 2006, Page(s):41 - 46
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6197 KB) | HTML iconHTML

    This paper describes a least absolute bound approach as a way to solve the ICA problems proposed in the 2006 MSLP competition. The least absolute bound is an ICA contrast closely related to the support width measure, which has been already studied for the blind extraction of bounded sources. By comparison, the least absolute bound applies to a broader class of sources, including those that are bou... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Joint Probabilistic-Deterministic Approach using Source-Filter Modeling of Speech Signal for Single Channel Speech Separation

    Publication Year: 2006, Page(s):47 - 52
    Cited by:  Papers (2)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5900 KB) | HTML iconHTML

    In this paper, we present a new technique for separating two speech signals from a single recording. For this purpose, we decompose the speech signal into the excitation signal and the vocal tract function and then estimate the components from the mixed speech using a hybrid model. We first express the probability density function (PDF) of the mixed speech's log spectral vectors in terms of the PD... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Wavelet Based Nonlinear Separation of Images

    Publication Year: 2006, Page(s):53 - 58
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5595 KB) | HTML iconHTML

    This work addresses a real-life problem corresponding to the separation of the nonlinear mixture of images which arises when we scan a paper document and the image from the back page shows through. The proposed solution consists of a non-iterative procedure that is based on two simple observations: (1) the high frequency content of images is sparse, and (2) the image printed on each side of the pa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Flexible ICA in Complex and Nonlinear Environment by Mutual Information Minimization

    Publication Year: 2006, Page(s):59 - 63
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5059 KB) | HTML iconHTML

    This paper introduces an Independent Component Analysis (ICA) approach to the separation of nonlinear mixtures in the complex domain. Source separation is performed by the minimization of output mutual information (MMI approach). Nonlinear complex functions involved in the processing are realized by the so called "splitting functions" which work on the real and the imaginary part of the signal res... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Separating Nonlinear Image Mixtures using a Physical Model Trained with ICA

    Publication Year: 2006, Page(s):65 - 70
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6035 KB) | HTML iconHTML

    This work addresses the separation of real-life nonlinear mixtures of images which occur when a paper document is scanned and the image from the back page shows through. We present a physical model of the mixing process, based on the consideration of the halftoning process used to print grayscale images. The corresponding inverse model is then used to perform image separation. The parameters of th... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Blind Signal Separation II

    Publication Year: 2006, Page(s): 71
    Request permission for commercial reuse | PDF file iconPDF (160 KB)
    Freely Available from IEEE
  • Blind Separation of Positive Dependent Sources by Non-Negative Least-Correlated Component Analysis

    Publication Year: 2006, Page(s):73 - 78
    Cited by:  Papers (7)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5748 KB) | HTML iconHTML

    Most independent component analysis methods for blind source separation rely on the fundamental assumption that all the unknown sources are mutually statistically independent. Such assumption becomes problematic when applied to many real world applications (e.g., biomedical imaging) that subsequently motivated the exploitation of non-negative nature of the sources, observations and mixing matrix. ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptable Nonlinearity for Complex Maximization of Nongaussianity and a Fixed-Point Algorithm

    Publication Year: 2006, Page(s):79 - 84
    Cited by:  Papers (18)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6258 KB) | HTML iconHTML

    Complex maximization of nonGaussianity (CMN) has been shown to provide reliable separation of both circular and non-circular sources using a class of complex functions in the non-linearity. In this paper, we derive a fixed-point algorithm for blind separation of noncircular sources using CMN. We also introduce the adaptive CMN (A-CMN) algorithm that provides significant performance improvement by ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Gradient and Fixed-Point Complex ICA Algorithms Based on Kurtosis Maximization

    Publication Year: 2006, Page(s):85 - 90
    Cited by:  Papers (10)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6735 KB) | HTML iconHTML

    We present two algorithms for independent component analysis of complex-valued signals based on the maximization of absolute value of kurtosis and establish their properties. Both the algorithm derivation and the analysis are carried out directly in the complex domain, without the use of complex-to-real mappings as the cost function satisfies Brandwood's analyticity condition. Simulation results a... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Towards Adaptive Blind Extraction of Post-Nonlinearly Mixed Signals

    Publication Year: 2006, Page(s):91 - 96
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (6629 KB) | HTML iconHTML

    A novel approach which extends blind source extraction (BSE) of one or group of sources to the case of post-nonlinear mixtures is proposed. This is achieved by an adaptive algorithm in which the cost function jointly estimates the kurtosis and a measure of nonlinearity. The analysis of both the quantitative and qualitative performance is provided, and simulation results are presented which illustr... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Subspace-Based Blind Identification of IIR Systems

    Publication Year: 2006, Page(s):97 - 102
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5434 KB) | HTML iconHTML

    A new subspace method for the blind identification of infinite impulse response (IIR), single input-multiple output (SIMO) systems represented using orthonormal bases with fixed poles, is presented in this paper. Basis coefficients are estimated in closed form, up to a scalar factor, by first computing the column space of the output Hankel matrix using singular value decomposition (SVD), and then ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Underdetermined Blind Source Separation Based on Generalized Gaussian Distribution

    Publication Year: 2006, Page(s):103 - 108
    Cited by:  Papers (5)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (5194 KB) | HTML iconHTML

    In this paper, a novel method for separating underlying sources with both sub- and super-Gaussian distributions from the underdetermined mixtures is proposed. The generalized Gaussian distribution (GGD) is used to model simultaneously both sub- and super-Gaussian distributions. The process of finding the most probable decomposition of the mixtures based on the GGD leads to that of minimizing the L... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Bayesian Learning and Modeling

    Publication Year: 2006, Page(s): 109
    Request permission for commercial reuse | PDF file iconPDF (166 KB)
    Freely Available from IEEE