Estimation of cosmological parameters using adaptive importance sampling (2009)
Wraith, Darren, Kilbinger, Martin, Benabed, Karim, Cappé, Olivier, Cardoso, Jean-François, Fort, Gersende, ...
We present a Bayesian sampling algorithm called adaptive importance sampling or Population Monte Carlo (PMC), whose computational workload is easily parallelizable and thus has the potential to...
Measuring the tensor to scalar ratio from CMB B-modes in presence of foregrounds (2009)
Betoule, Marc, Pierpaoli, E., Delabrouille, J., Jeune, M. Le, Cardoso, Jean-François
Abreg: We investigate the impact of polarized foreground emission on the performances of future CMB experiments in measuring the tensor-to-scalar ratio r. We design a component separation pipeline,...
Measuring the tensor to scalar ratio from CMB B-modes in presence of foregrounds (2009)
Betoule, Marc, Pierpaoli, E., Delabrouille, J., Le Jeune, M., Cardoso, Jean-François
Abreg: We investigate the impact of polarized foreground emission on the performances of future CMB experiments in measuring the tensor-to-scalar ratio r. We design a component separation pipeline,...
Measuring the tensor to scalar ratio from CMB B-modes in presence of foregrounds (2009)
Betoule, Marc, Pierpaoli, E., Delabrouille, J., Le Jeune, M., Cardoso, Jean-François
Abreg: We investigate the impact of polarized foreground emission on the performances of future CMB experiments in measuring the tensor-to-scalar ratio r. We design a component separation pipeline,...
Measuring the tensor to scalar ratio from CMB B-modes in presence of foregrounds (2009)
Betoule, Marc, Pierpaoli, E., Delabrouille, J., Le Jeune, M., Cardoso, Jean-François
Abreg: We investigate the impact of polarized foreground emission on the performances of future CMB experiments in measuring the tensor-to-scalar ratio r. We design a component separation pipeline,...
Dependence, Correlation and Gaussianity in Independent Component Analysis (2008)
Independent component analysis (ICA) is the decomposition of a random vector in linear components which are “as independent as possible. ” Here, “independence ” should be understood in its...
Overlearning in Marginal Distribution Based ICA: Analysis and Solutions (2008)
Jaakko Särelä, Ricardo Vigário, Fraunhofer First. Ida, Te-won Lee, Jean-françois Cardoso, Erkki Oja, ...
The present paper is written as a word of caution, with users of independent component analysis (ICA) in mind, to overlearning phenomena that are often observed. We consider two types of...
Séparation Adaptative De Sources En Aveugle. Implantation Complexe Sans Contraintes. (2008)
Beate Christina, Jean-François Cardoso
A new adaptive blind algorithm for the separation of ind'ependent complex (or real) sources is presented. The algorithm stems from a classical cumulant-based separation criterion, but is not...
Separation of Non Stationary Sources; Achievable Performance (2007)
We consider the blind separation of an instantaneous mixture of non stationary source signals, possibly normally distributed. The asymptotic Cramer-Rao bound is exhibited in the case of known source...
Séparation Adaptative De Sources Dans L'espace Signal. (2007)
To solve the source separation problem, it is possible to take advantage of `extra' sensors to reduce the effect of noise by projecting the observations onto the `signal subspace'. In this...
SEPARATION OF NON STATIONARY SOURCES; ACHIEVABLE PERFORMANCE (2007)
We consider the blind separation of an instantaneous mixture of non stationary source signals, possibly normally distributed. The asymptotic Cramér-Rao bound is exhibited in the case of known source...
This communication deals with higher-order multivariate statistics. A first theoretical part is followed by an original application. In the first part we propose a special (index-free) tensor...
Iterative Techniques For, Using Only, Fourth-order Cumulants, Jean-françois Cardoso
Blind source separation" is an array processing problem without a priori information (no array manifold). This model can be identified resorting to 4th-order cumulants only via the concept of...
Direction Finding Algorithms, Eric Moulines, Jean-françois Cardoso
This communication deals with high-resolution direction finding using higher-order cumulants of the array data. Two 4th-order cumulant-based matrices are considered : the diagonal cumulant slice and...
Looking for components in the Universe's oldest data set (2007)
Jean-françois Cardoso, Centre National, Recherche Scientifique, École Nationale, Supérieure Télécommunications
Modern astronomical experiments o#er many exciting challenges in the field of statistical signal processing. This paper gives an introduction to the problem of imaging the cosmic microwave background...
Component Separation For Cosmic Microwave Background Data: A Blind (2007)
Approach Based On, Guillaume Patanchon, Hichem Snoussi, Jean-françois Cardoso, Jacques Delabrouille
We present a blind multi-detector multi-component spectral matching method for all sky observations of the cosmic microwave background, working on the spherical harmonics basis. The method allows to...
identification aveugle, s'eparation de sources. Keywords: Higher-order statistics, blind identification, source separation. Remerciements `a Jean-Fran¸cois Bercher pour avoir...
Practical wavelet design on the sphere (2007)
Guilloux, Frédéric, Fay, Gilles, Cardoso, Jean-François
We address the question of designing isotropic analysis functions on the sphere which are perfectly limited in the spectral domain and optimally localized in the spatial domain. This work is...
Practical wavelet design on the sphere (2007)
Guilloux, Frédéric, Fay, Gilles, Cardoso, Jean-François
We address the question of designing isotropic analysis functions on the sphere which are perfectly limited in the spectral domain and optimally localized in the spatial domain. This work is...
Practical wavelet design on the sphere (2007)
Guilloux, Frédéric, Fay, Gilles, Cardoso, Jean-François
We address the question of designing isotropic analysis functions on the sphere which are perfectly limited in the spectral domain and optimally localized in the spatial domain. This work is...
Practical wavelet design on the sphere (2007)
Guilloux, Frédéric, Fay, Gilles, Cardoso, Jean-François
We address the question of designing isotropic analysis functions on the sphere which are perfectly limited in the spectral domain and optimally localized in the spatial domain. This work is...
Energy-based models for sparse overcomplete representations (2003)
Yee Whye Teh, Max Welling, Simon Osindero, Geoffrey E. Hinton, Te-won Lee, Jean-françois Cardoso, ...
We present a new way of extending independent components analysis (ICA) to overcomplete representations. In contrast to the causal generative extensions of ICA which maintain marginal independence of...
Andreas Ziehe, Motoaki Kawanabe, Stefan Harmeling, Klaus-Robert Müller, Fraunhofer First. Ida, Te-won Lee, ...
We propose two methods that reduce the post-nonlinear blind source separation problem (PNLBSS) to a linear BSS problem. The first method is based on the concept of maximal correlation: we apply the...
ICA Using Spacings Estimates of Entropy (2003)
Te-won Lee, Jean-françois Cardoso, Erkki Oja, Shun-ichi Amari
This paper presents a new algorithm for the independent components analysis (ICA) problem based on an efficient entropy estimator. Like many previous methods, this algorithm directly minimizes the...
Independent Component Analysis (2003)
Of The Cosmic, Jean-françois Cardoso, Jacques Delabrouille, Guillaume Patanchon
This paper presents an application of ICA to astronomical imaging. A first section describes the astrophysical context and motivates the use of source separation ideas. A second section describes our...
ICA for watermarking digital images (2003)
Stéphane Bounkong, Borémi Toch, David Saad, David Lowe, Te-won Lee, Jean-françois Cardoso, ...
We present a domain-independent ICA-based approach to watermarking. This approach can be used on images, music or video to embed either a robust or fragile watermark. In the case of robust...
A maximum likelihood approach to single-channel source separation (2003)
Gil-jin Jang, Te-won Lee, Te-won Lee, Jean-françois Cardoso, Erkki Oja, Shun-ichi Amari
This paper presents a new technique for achieving blind signal separation when given only a single channel recording. The main concept is based on exploiting a priori sets of time-domain basis...
A multiscale framework for blind separation of linearly mixed signals (2003)
Pavel Kisilev, Michael Zibulevsky, Yehoshua Y. Zeevi, Te-won Lee, Jean-françois Cardoso, Erkki Oja, ...
We consider the problem of blind separation of unknown source signals or images from a given set of their linear mixtures. It was discovered recently that exploiting the sparsity of sources and their...
Energy-based models for sparse overcomplete representations (2003)
Yee Whye Teh, Max Welling, Simon Osindero, Geoffrey E. Hinton, Te-won Lee, Jean-françois Cardoso, ...
We present a new way of extending independent components analysis (ICA) to overcomplete representations. In contrast to the causal generative extensions of ICA which maintain marginal independence of...
The three easy routes to independent component analysis; contrasts and geometry (2001)
Blind separation of independent sources can be achieved by exploiting non Gaussianity, non stationarity or time correlation. This paper examines in a unified framework the objective functions...
The Three Easy Routes To Independent Component Analysis; Contrasts And Geometry. (2001)
Blind separation of independent sources can be achieved by exploiting non Gaussianity, non stationarity or time correlation. This paper examines in a unified framework the objective functions...
Blind Separation of Instantaneous Mixtures of Non Stationary Sources (2000)
Dinh-tuan Pham, Jean-François Cardoso
Most ICA algorithms are based on a model of stationary sources. This paper considers exploiting the (possible) non-stationarity of the sources to achieve separation. We introduce two objective...
Invariance of Subspace Based Estimators (2000)
Jean-françois Cardoso, Éric Moulines
Subspace based estimates, i.e. estimates obtained by exploiting the orthogonality between a sample subspace and a parameter-dependent subspace have proved useful in many applications, including array...
Some experiments on independent component analysis of non-Gaussian processes. (1999)
Jean-François Cardoso, David L. Donoho
This paper reports on numerical experiments on the `independent component analysis' (ICA) of some nonGaussian stochastic processes. It is found that the orthonormal basis discovered by ICA are...
A Robustness Property of DOA Estimators Based on Covariance (1999)
Jean-François Cardoso, Éric Moulines
A simple general formula is derived under a Gaussian model for the asymptotic covariance of direction-of-arrival (DOA) estimators based on the covariance of the sensor array data. It is then...
Some Experiments on Independent Component Analysis (1999)
This paper reports on numerical experiments on the `independent component analysis' (ICA) of some nonGaussian stochastic processes. It is found that the orthonormal basis discovered by ICA are...
On the Stability of Source Separation Algorithms (1998)
. This paper discusses the stability of algorithms for independent component analysis and the blind separation of signals. It builds on previous analysis of local asymptotic stability to present a...
Blind Signal Separation: Statistical Principles (1998)
Blind signal separation (BSS) and independent component analysis (ICA) are emerging techniques of array processing and data analysis, aiming at recovering unobserved signals or `sources' from...
Learning in Manifolds: The Case of Source Separation (1998)
The blind signal separation (BSS) problem has a distinctive feature: the unknown parameter being an invertible matrix, the parameter set is a multiplicative group and the observations can be modeled...
Multidimensional Independent Component Analysis. (1998)
This discussion paper proposes to generalize the notion of Independent Component Analysis (ICA) to the notion of Multidimensional Independent Component Analysis (MICA). We start from the ICA or blind...
Éric Moulines, Jean-François Cardoso, Elisabeth Gassiat
In this paper, an approximate maximum likelihood method for blind source separation and deconvolution of noisy signal is proposed. This technique relies upon a data augmentation scheme, where the...
Estimating Equations For Source Separation (1997)
This paper proposes a unifying view of source separation via the concepts of `estimating function' and `estimating equation'. We exhibit the estimating functions corresponding to various...
Infomax and Maximum Likelihood for Blind Source Separation (1997)
Algorithms for the blind separation of sources can be derived from several different principles. This letter shows that the recently proposed infomax principle is equivalent to maximum likelihood....
A Blind Source Separation Technique Using Second Order Statistics (1997)
Adel Belouchrani, Karim Abed-meraim, Jean-François Cardoso, Eric Moulines, Ieee Member, Ieee Member, ...
Separation of sources consists in recovering a set of signals of which only instantaneous linear mixtures are observed. In many situations, no a priori information on the mixing matrix is available:...
Equivariant Blind Deconvolution of MIMO-FIR Channels (1997)
Alexei Gorokhov, Jean-François Cardoso
We address the blind identi cation and deconvolution of Multiple Input Multiple Output (MIMO) linear FIR channels. This is an instance of blind separation of convolutive mixtures. The unknown system...
Invariance of Subspace Based Estimators (1996)
Jean-François Cardoso, Éric Moulines
Subspace based estimates, i.e. estimates obtained by exploiting the orthogonality between a sample subspace and a parameter-dependent subspace have proved useful in many applications, including array...
Equivariant Adaptive Source Separation (1996)
Jean-François Cardoso, Beate Laheld
Source separation consists in recovering a set of independent signals when only mixtures with unknown coefficients are observed. This paper introduces a class of adaptive algorithms for source...
Independent Component Analysis, A Survey Of Some Algebraic Methods. (1996)
Jean-François Cardoso, Pierre Comon
The source separation problem has been addressed in many ways during the last decade, and one of its instances gave birth to Independent Component Analysis (ICA). Iterative methods can be opposed to...
Equivariant Adaptive Source Separation (1996)
Jean-François Cardoso, Beate Hvam Laheld
Source separation consists in recovering a set of independent signals when only mixtures with unknown coefficients are observed. This paper introduces a class of adaptive algorithms for source...
The Invariant Approach to Source Separation (1995)
The notion of equivariance is relevant to source separation because multiplication of mixed signals is equivalent to changing the unknown parameter (the mixing matrix) into another mixing matrix....
Adel Belouchrani, Jean-François Cardoso, Jean-fran��cois Cardoso
This paper deals with the source separation problem which consists in the separation of a mixture of independent sources without a priori knowledge on the mixing matrix. When the source distributions...
Perturbation of joint diagonalizers - Perturbation de la diagonalisation conjointe (1995)
. This report gives the first order perturbation of the joint diagonalization of a set of commuting matrices. R'esum'e. Ce rapport donne la perturbation au premier ordre de la...
A Tetradic Decomposition of 4th-Order Tensors. Application to the Source Separation Problem (1995)
. Two results are presented on a SVD-like decomposition of 4th-order tensors. This is motivated by an array processing problem: consider an array of m sensors listening at n independent narrow band...
Output Cumulant Matching For Source Separation (1995)
Jean-François Cardoso, Sandip Bose, Benjamin Friedlander
Cumulant-based criteria or contrast functions are considered for solving the source separation problem. This paper investigates the optimal use of the whole set of 4th-order cumulants. An optimal...
Maximum Likelihood Source Separation for Discrete Sources (1994)
Adel Belouchrani, Jean-François Cardoso
. This communication deals with the source separation problem which consists in the separation of a noisy mixture of independent sources without a priori knowledge of the mixture coefficients. In...
How Much More Doa Information In Higher-Order Statistics? (1994)
Jean-François Cardoso, Éric Moulines
We consider the use of 2nd- and 4th-order cumulants for estimating the direction-of-arrival (DOAs) in narrow band array processing. The Fisher information about the DOAs contained in several cumulant...
Adaptive Source Separation With Uniform Performance (1994)
Beate Laheld, Jean-François Cardoso
. This paper presents a family of adaptive algorithms for the blind separation of independent signals. Source separation consists in recovering a set of independent signals from some linear mixtures...
On the Performance of Orthogonal Source Separation Algorithms (1994)
. Source separation consists in recovering a set of n independent signals from m n observed instantaneous mixtures of these signals, possibly corrupted by additive noise. Many source separation...
An Efficient Technique For The Blind Separation Of Complex Sources. (1993)
Jean-François Cardoso, Antoine Souloumiac
Blind identification of spatial mixtures allows an array of sensors to implement source separation when the array manifold is unknown. A family of 4th-order cumulant-based criteria for blind source...
Blind Beamforming for Non Gaussian Signals (1993)
Jean-François Cardoso, Antoine Souloumiac
This paper considers an application of blind identification to beamforming. The key point is to use estimates of directional vectors rather than resorting to their hypothesized value. By using...
Fourth-Order Cumulant Structure Forcing. Application to Blind Array Processing (1992)
In blind array processing, the array manifold is unknown but, under the signal independence assumption, the signal parameters can be estimated by resorting to higher-order information. We consider...
Iterative Techniques For Blind Source Separation Using Only Fourth-Order Cumulants (1992)
Using Only, Fourth-order Cumulants, Jean-François Cardoso
. "Blind source separation" is an array processing problem without a priori information (no array manifold). This model can be identified resorting to 4th-order cumulants only via the...
Direction Finding Algorithms Using Fourth Order Statistics. Asymptotic Performance Analysis (1992)
Eric Moulines, Jean-François Cardoso
This communication deals with high-resolution direction finding using higher-order cumulants of the array data. Two 4th-order cumulant-based matrices are considered : the diagonal cumulant slice and...
Comparaison de m'ethodes de s'eparation de sources (1991)
Antoine Souloumiac, Jean-françois Cardoso
Nous présentons dans cette communication une nouvelle classe de méthodes permettant de séparer un mélange linéaire bruité de sources indépendantes, en traitement d’antenne bande étroite....
Comparaisons De Methodes De Separation De Sources (1991)
Antoine Souloumiac, Jean-François Cardoso
We present in this communication a new class of methods aimed at retrieving independent components from a linear mixture in narrow-band array processing context. These methods are based on the joint...
Second-Order Versus Fourth-Order Music Algorithms: An Asymptotical Statistical Analysis (1991)
Eric Moulines, Jean-François Cardoso
Direction finding techniques are usually based on the 2ndorder statistics of the received data. In this paper, we propose a MUSIC-like direction finding algorithm which uses a matrix-valued statistic...
Higher-Order Narrow-Band Array Processing (1991)
This communication deals with narrow-band array processing using higher-order cumulants. A versatile tensor formalism is presented and adopted throughout to handle higher-order statistics of complex...
This communication deals with higher-order multivariate statistics. A first theoretical part is followed by an original application. In the first part we propose a special (index-free) tensor...
Source Separation Using Higher Order Moments (1989)
This communication presents a simple algebraic method for the extraction of independent components in multidimensional data. Since statistical independence is a much stronger property than...
Overlearning in marginal distribution-based ICA: analysis and solutions (1447)
Jaakko Särelä, Ricardo Vigário, Fraunhofer First. Ida, Te-won Lee, Jean-françois Cardoso, Erkki Oja, ...
The present paper is written as a word of caution, with users of independent component analysis (ICA) in mind, to overlearning phenomena that are often observed. We consider two types of...