Width optimization of the Gaussian kernels in radial basis function networks (2010)
Nabil Benoudjit, Cédric Archambeau, Amaury Lendasse, John Lee, Michel Verleysen
Abstract. Radial basis function networks are usually trained according to a three-stage procedure. In the literature, many papers are devoted to the estimation of the position of Gaussian kernels, as...
Self Organizing Star (SOS) for health monitoring (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Self Organizing Maps (SOM) have been successfully applied in a lot of real world hard problems since their apparition. In this paper we present new topologies for SOM based on a planar graph. The...
Self Organizing Star (SOS) for health monitoring (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Self Organizing Maps (SOM) have been successfully applied in a lot of real world hard problems since their apparition. In this paper we present new topologies for SOM based on a planar graph. The...
Aircraft engine health monitoring using Self-Organizing Maps (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Aircraft engines are designed to be used during several tens of years. Ensuring a proper operation of engines over their lifetime is therefore an important and difficult task. The maintenance can be...
Trajectory Clustering for Vibration Detection in Aircraft Engines (2010)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
The automatic detection of the vibration signature of rotating parts of an aircraft engine is considered. This paper introduces an algorithm that takes into account the variation over time of the...
Aircraft engine health monitoring using Self-Organizing Maps (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Aircraft engines are designed to be used during several tens of years. Ensuring a proper operation of engines over their lifetime is therefore an important and difficult task. The maintenance can be...
Trajectory Clustering for Vibration Detection in Aircraft Engines (2010)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
The automatic detection of the vibration signature of rotating parts of an aircraft engine is considered. This paper introduces an algorithm that takes into account the variation over time of the...
Aircraft engine health monitoring using Self-Organizing Maps (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Aircraft engines are designed to be used during several tens of years. Ensuring a proper operation of engines over their lifetime is therefore an important and difficult task. The maintenance can be...
Trajectory Clustering for Vibration Detection in Aircraft Engines (2010)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
The automatic detection of the vibration signature of rotating parts of an aircraft engine is considered. This paper introduces an algorithm that takes into account the variation over time of the...
Self Organizing Star (SOS) for health monitoring (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Self Organizing Maps (SOM) have been successfully applied in a lot of real world hard problems since their apparition. In this paper we present new topologies for SOM based on a planar graph. The...
Aircraft engine health monitoring using Self-Organizing Maps (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Aircraft engines are designed to be used during several tens of years. Ensuring a proper operation of engines over their lifetime is therefore an important and difficult task. The maintenance can be...
Trajectory Clustering for Vibration Detection in Aircraft Engines (2010)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
The automatic detection of the vibration signature of rotating parts of an aircraft engine is considered. This paper introduces an algorithm that takes into account the variation over time of the...
Self Organizing Star (SOS) for health monitoring (2010)
Côme, Etienne, Cottrell, Marie, Verleysen, Michel, Lacaille, Jérôme
Self Organizing Maps (SOM) have been successfully applied in a lot of real world hard problems since their apparition. In this paper we present new topologies for SOM based on a planar graph. The...
Université catholique de Louvain, Machine Learning Group, (2009)
Michel Verleysen, Fabrice Rossi
Abstract. The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps...
Rui Nian, Guangrong Ji, Michel Verleysen
This paper presents a scheme for unsupervised classification with Gaussian mixture models by means of statistical learning analysis. A Bayesian Ying-Yang harmony learning system acts as a statistical...
An Alternative to Center-Based Clustering Algorithm Via Statistical Learning Analysis (2009)
Rui Nian, Guangrong Ji, Michel Verleysen
Abstract. This paper presents an alternative for center-based clustering algorithms, in particular the k-means algorithm, via statistical learning analysis. The essence of statistical learning...
Alex Assenza, Maurizio Valle, Michel Verleysen
Probability density estimation (PDF) is a task of primary importance in many contexts, including Bayesian learning and novelty detection. Despite the wide variety of methods at disposal to estimate...
On the Risk of Using Rényi’s Entropy for Blind Source Separation (2009)
Dinh-tuan Pham, Frédéric Vrins, Graduate Student Member, Michel Verleysen, Senior Member
Abstract—Recently, some researchers have suggested Rényi’s entropy in its general form as a blind source separation (BSS) objective function. This was motivated by two arguments: 1) Shannon’s...
K-Nearest Neighbours based on Mutual Information for Incomplete Data Classification (2009)
Pedro J. García-laencina, José-luis Sancho-gómez, Anibal R. Figueiras-vidal, Michel Verleysen
Abstract. Incomplete data is a common drawback that machine learning techniques need to deal with when solving real-life classification tasks. One of the most popular procedures for solving this kind...
1- Université catholique de Louvain- Machine Learning Group (2009)
Victor Onclinx, Vincent Wertz, Michel Verleysen
Nonlinear data projection on a sphere with controlled trade-off between trustworthiness and continuity
Improving the Robustness to Outliers of Mixtures of Probabilistic PCAs (2009)
Nicolas Delannay, Cédric Archambeau, Michel Verleysen
Abstract. Principal Component Analysis, when formulated as a probabilistic model, can be made robust to outliers by using a Student-t assumption on the noise distribution instead of a Gaussian one....
Using the Delta Test for Variable Selection (2009)
Emil Eirola, Elia Liitiäinen, Amaury Lendasse, Francesco Corona, Michel Verleysen
Abstract. Input selection is an important consideration in all large-scale modelling problems. We propose that using an established noise variance estimator known as the Delta test as the target to...
Residual variance estimation in machine learning (2009)
Liitiainen, Elia, Verleysen, Michel, Corona, Francesco, Lendasse, Amaury
The problem of residual variance estimation consists of estimating the best possible generalization error obtainable by any model based on a finite sample of data. Even though it is a natural...
Advances in Feature Selection with Mutual Information (2009)
Verleysen, Michel, Rossi, Fabrice, François, Damien
The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps fighting the...
Fault prediction in aircraft engines using Self-Organizing Maps (2009)
Cottrell, Marie, Gaubert, Patrice, Eloy, Cédric, François, Damien, Hallaux, Geoffroy, Lacaille, Jérôme, ...
Aircraft engines are designed to be used during several tens of years. Their maintenance is a challenging and costly task, for obvious security reasons. The goal is to ensure a proper operation of...
An Empirical Taxonomy of Start-Up Firms Growth Trajectories (2009)
Mahamadou Biga Diambeidou, Damien François, Benoît Gailly, Michel Verleysen
Over the past decades, new and small firm growth has received considerable attention from researchers and policy-makers around the world. New firms have been identified
Mixtures of Robust Probabilistic Principal Component Analyzers (2009)
Cédric Archambeau, Nicolas Delannay, Michel Verleysen
Abstract. Discovering low-dimensional (nonlinear) manifolds is an important problem in Machine Learning. In many applications, the data are in a high dimensional space. This can be problematic in...
Residual Variance Estimation in Machine Learning (2009)
Liitiäinen, Elia, Verleysen, Michel, Corona, Francesco, Lendasse, Amaury
The problem of residual variance estimation consists of estimating the best possible generalization error obtainable by any model based on a finite sample of data. Even though it is a natural...
Fault prediction in aircraft engines using Self-Organizing Maps (2009)
Cottrell, Marie, Gaubert, Patrice, Eloy, Cédric, François, Damien, Hallaux, Geoffroy, Lacaille, Jérôme, ...
Aircraft engines are designed to be used during several tens of years. Their maintenance is a challenging and costly task, for obvious security reasons. The goal is to ensure a proper operation of...
Fault prediction in aircraft engines using Self-Organizing Maps (2009)
Cottrell, Marie, Gaubert, Patrice, Eloy, Cédric, François, Damien, Hallaux, Geoffroy, Lacaille, Jérôme, ...
Aircraft engines are designed to be used during several tens of years. Their maintenance is a challenging and costly task, for obvious security reasons. The goal is to ensure a proper operation of...
Advances in Feature Selection with Mutual Information (2009)
Verleysen, Michel, Rossi, Fabrice, François, Damien
The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps fighting the...
Advances in Feature Selection with Mutual Information (2009)
Verleysen, Michel, Rossi, Fabrice, François, Damien
The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps fighting the...
Random model of vibrations for Foreign Object Damage detection in a civil aircraft engine (2009)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
In this article we analyze several vibration time series measured on a real fan test rig before and after it is hit by a flying object. We show first evidence that a windowed autoregressive model may...
Random model of vibrations for Foreign Object Damage detection in a civil aircraft engine (2009)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
In this article we analyze several vibration time series measured on a real fan test rig before and after it is hit by a flying object. We show first evidence that a windowed autoregressive model may...
Random model of vibrations for Foreign Object Damage detection in a civil aircraft engine (2009)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
In this article we analyze several vibration time series measured on a real fan test rig before and after it is hit by a flying object. We show first evidence that a windowed autoregressive model may...
Fault prediction in aircraft engines using Self-Organizing Maps (2009)
Cottrell, Marie, Gaubert, Patrice, Eloy, Cédric, François, Damien, Hallaux, Geoffroy, Lacaille, Jérôme, ...
Aircraft engines are designed to be used during several tens of years. Their maintenance is a challenging and costly task, for obvious security reasons. The goal is to ensure a proper operation of...
Advances in Feature Selection with Mutual Information (2009)
Verleysen, Michel, Rossi, Fabrice, François, Damien
The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps fighting the...
Random model of vibrations for Foreign Object Damage detection in a civil aircraft engine (2009)
Hazan, Aurélien, Verleysen, Michel, Cottrell, Marie, Lacaille, Jérôme
In this article we analyze several vibration time series measured on a real fan test rig before and after it is hit by a flying object. We show first evidence that a windowed autoregressive model may...
Fault prediction in aircraft engines using Self-Organizing Maps (2009)
Cottrell, Marie, Gaubert, Patrice, Eloy, Cédric, François, Damien, Hallaux, Geoffroy, Lacaille, Jérôme, ...
Aircraft engines are designed to be used during several tens of years. Their maintenance is a challenging and costly task, for obvious security reasons. The goal is to ensure a proper operation of...
Advances in Feature Selection with Mutual Information (2009)
Verleysen, Michel, Rossi, Fabrice, François, Damien
The selection of features that are relevant for a prediction or classification problem is an important problem in many domains involving high-dimensional data. Selecting features helps fighting the...
Business Plans Classification with Locally Pruned Lazy Learning Models (2008)
Antti Sorjamaa, Amaury Lendasse, Damien Francois, Michel Verleysen
Un plan d’affaires est un document présentant de manière concise les éléments clefs qui décrivent un projet de création d’entreprise. Celui-ci est utilisé comme un outil parmi d’autres...
Fabrice Rossi, Damien François, Vincent Wertz, Michel Verleysen
A functional approach to variable selection in
Collaborative Filtering with interlaced Generalized Linear Models (2008)
Nicolas Delannay, Michel Verleysen
Abstract. Collaborative Filtering (CF) aims at finding patterns in a sparse matrix of contingency. It can be used for example to mine the ratings given by users on a set of items. In this paper, we...
† Université catholique de Louvain Machine Learning Group (2008)
Cédric Archambeau, Torsten Butz, Vlad Popovici, Michel Verleysen, Place Du Levant
In this paper, supervised nonparametric information theoretic classification (ITC) is introduced. Its principle relies on the likelihood of a data sample of transmitting its class label to data...
Lag selection for regression models using high-dimensional mutual information, in (2008)
Geoffroy Simon, Michel Verleysen
Abstract. Mutual information may be used to select the embedding lag of a time series. However, this lag selection is usually limited to the analysis of the mutual information between a pair of...
M.: Generalization of the Lp norm for time series and its application to self-organizing maps (2008)
Abstract- Time series are often encoded in vectors and analyzed using standard vectorial tools (distances, inner products, etc.). Most of them neglect the temporal structure of time series. This...
A Functional Approach to Variable Selection in Spectrometric Problems ⋆ (2008)
Fabrice Rossi, Vincent Wertz, Michel Verleysen, Chesnay Cedex
Abstract. In spectrometric problems, objects are characterized by highresolution spectra that correspond to hundreds to thousands of variables. In this context, even fast variable selection methods...
A Minimum-Range Approach to Blind Extraction of Bounded Sources (2008)
Frédéric Vrins, Student Member, John A. Lee, Michel Verleysen, Senior Member
Abstract—In spite of the numerous approaches that have been derived for solving the independent component analysis (ICA) problem, it is still interesting to develop new methods when, among other...
Zero-entropy minimization for blind extraction of bounded sources (BEBS (2008)
Frédéric Vrins, Deniz Erdogmus, Christian Jutten, Michel Verleysen
Abstract. Renyi’s entropy can be used as a cost function for blind source separation (BSS). Previous works have emphasized the advantage of setting Renyi’s exponent to a value different from one...
27 th August 9:00-10:00 Session 1 (2008)
Mahamadou Biga Diambeidou, Damien François, Benoît Gailly, Michel Verleysen, Vincent Wertz, Université Catholique De Louvain, ...
Deroy and N. Delobe for their comments. Empirical Taxonomy of Start-Up Firms Growth Trajectories This article provides a method that can accommodate, in a systematic way, the analysis of new firm...
Information-Theoretic Feature Selection for the Classification of Hysteresis Curves ⋆ (2008)
Vanessa Gómez-verdejo, Michel Verleysen, Jérôme Fleury
Abstract. This paper presents a methodology for functional data analysis. It consists in extracting a large number of features with maximal content of information and then selecting the appropriate...
Mixing and Non-Mixing Local Minima of the Entropy Contrast for Blind Source Separation (2008)
Frédéric Vrins, Student Member, Dinh-tuan Pham, Michel Verleysen, Senior Member
Abstract—In this paper, both non-mixing and mixing local minima of the entropy are analyzed from the viewpoint of blind source separation (BSS); they correspond respectively to acceptable and...
2 Machine Learning Group, (2008)
Amaury Lendasse, Yongnan Ji, Nima Reyhani, Michel Verleysen, Louvain-la-neuve Belgique
Abstract. This paper presents a new method for the selection of the two hyperparameters of Least Squares Support Vector Machine (LS-SVM) approximators with Gaussian Kernels. The two hyperparameters...
DD-HDS: A Method for Visualization and Exploration of High-Dimensional Data (2008)
Sylvain Lespinats, Michel Verleysen, Senior Member, Alain Giron, Bernard Fertil
Abstract—Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven...
ICP12 A Least Absolute Bound Approach to ICA - Application to the MLSP 2006 competition (2008)
John A. Lee, Frédéric Vrins, Michel Verleysen
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...
Automatic Adjustment of Discriminant Adaptive Nearest Neighbor (2008)
Nicolas Delannay, Cédric Archambeau, Michel Verleysen
K-Nearest Neighbors relies on the definition of a global metric. In contrast, Discriminant Adaptive Nearest Neighbor (DANN) computes a different metric at each query point based on a local Linear...
EMBEDDED FUZZY CONTROL FOR AUTOMATIC CHANNEL EQUALIZATION AFTER DIGITAL TRANSMISSIONS (2008)
Carlos Dualibe, Paul Jespers, Michel Verleysen
ABSTRACT: A straightforward technique for automatic adaptation of channels equalizers after digital data transmission is presented. Inter-Symbol Interference (ISI) at the received signal is...
Open Questions About Similarity Search in High-Dimensional Spaces (2008)
Damien Francois Vincent, Vincent Wertz, Michel Verleysen
he norm (or Euclidean distance to origin) of a random vector gets small compared to the expected value of the norm. This means that the norm of a high dimensional vector becomes nearly a constant...
with Locally Pruned Lazy Learning Models (2008)
Antti Sorjamaa Amaury, Amaury Lendasse, Damien Francois, Michel Verleysen
describing a perceived business opportunity. It is used among others as a tool for evaluating the feasibility and profitability of a project. In this paper, we study the relationship between business...
A Current-Mode Cmos Loser-Take-All With (2008)
Minimum Function For, Nicolas Donckers, Carlos Dualibe, Michel Verleysen
A novel architecture for loser-take-all functions is proposed. Inputs and outputs of the circuit are currents, which make the circuit appropriated for low-voltage neural hardware computation. In...
Manifold Constrained Variational Mixtures (2008)
Cedric Archambeau And, Cédric Archambeau, Michel Verleysen
In many data mining applications, the data manifold is of lower dimension than the dimension of the input space. In this paper, it is proposed to take advantage of this additional information in the...
Entropy Minima and Distribution Structural (2008)
Modifications In Blind, Frédéric Vrins, Cédric Archambeau, Michel Verleysen
The source separation problem is usually solved through a gradient descent on a cost function C . However, C may have local minima that are irrelevant from the source separation point of view in...
About the Locality of Kernels in (2008)
High-Dimensional Spaces Damien, Damien Francois, Vincent Wertz, Michel Verleysen
Gaussian kernels are widely used in many data analysis tools such as Radial-Basis Function networks, Support Vector Machines and many others. Gaussian kernels are most often deemed to provide a local...
Non-linear ICA by Using Isometric (2008)
Dimensionality Reduction John, John A. Lee, Christian Jutten, Michel Verleysen
In usual ICA methods, sources are typically estimated by maximizing a measure of their statistical independence. This paper explains how to perform non-linear ICA by preprocessing the mixtures with...
Supervised Nonparametric Information Theoretic Classification (2008)
Edric Archambeau Torsten, Torsten Butz, Vlad Popovici, Michel Verleysen, Place Du Levant
theoretic classification (ITC) is introduced. Its principle relies on the likelihood of a data sample of transmitting its class label to data points in its vicinity. ITC's learning rule is...
Manifold Constrained Finite Gaussian Mixtures (2008)
Cedric Archambeau And, Cédric Archambeau, Michel Verleysen
In many practical applications, the data is organized along a manifold of lower dimension than the dimension of the embedding space.
Krier, Catherine, Rossi, Fabrice, François, Damien, Verleysen, Michel
Prediction problems from spectra are largely encountered in chemometry. In addition to accurate predictions, it is often needed to extract information about which wavelengths in the spectra...
Krier, Catherine, Rossi, Fabrice, François, Damien, Verleysen, Michel
Prediction problems from spectra are largely encountered in chemometry. In addition to accurate predictions, it is often needed to extract information about which wavelengths in the spectra...
Krier, Catherine, Rossi, Fabrice, François, Damien, Verleysen, Michel
Prediction problems from spectra are largely encountered in chemometry. In addition to accurate predictions, it is often needed to extract information about which wavelengths in the spectra...
Using the Delta test for variable selection (2008)
Eirola, Emil, Liitiäinen, Elia, Lendasse, Amaury, Corona, Francesco, Verleysen, Michel
We propose that using an established noise variance estimator known as the Delta test as the target to minimise can provide an effective input selection methodology. Theoretical justifications and...
Collaborative filtering with interlaced generalized linear models (2008)
Nicolas Delannay, Michel Verleysen
Collaborative filtering (CF) is a data analysis task appearing in many challenging applications, in particular data mining in Internet and e-commerce. CF can often be formulated as identifying...
Krier, Catherine, Rossi, Fabrice, François, Damien, Verleysen, Michel
Prediction problems from spectra are largely encountered in chemometry. In addition to accurate predictions, it is often needed to extract information about which wavelengths in the spectra...
Krier, Catherine, Rossi, Fabrice, François, Damien, Verleysen, Michel
Prediction problems from spectra are largely encountered in chemometry. In addition to accurate predictions, it is often needed to extract information about which wavelengths in the spectra...
Double SOM for long-term time series prediction (2007)
Geoffroy Simon, Amaury Lendasse, Marie Cottrell, Jean-claude Fort, Michel Verleysen
Abstract--- Many time series forecasting problems require the estimation of possibly inaccurate, but longterm, trends, rather than accurate short-term prediction. In this paper, a double use of the...
Many engineering problems include some kind of recognition: from automatic character recognition to the control of steel quality in a steelworks, through the fault detection in nuclear plants or the...
Hopfield’s This survey of (2007)
Michel Verleysen, Paul G. A. Jespers
the pros and cons of analog devices presents a 14-neuron test chip and an algorithm for fully interconnected networks.
2002 Special Issue Self-organizing maps with recursive neighborhood adaptation (2007)
Self-organizing maps (SOMs) are widely used in several fields of application, from neurobiology to multivariate data analysis. In that context, this paper presents variants of the classic SOM...
Width optimization of the Gaussian kernels in radial basis function networks (2007)
Nabil Benoudjit, Cédric Archambeau, Amaury Lendasse, John Lee, Michel Verleysen, Avenue G. Lemaître, ...
Abstract. Radial basis function networks are usually trained according to a three-stage procedure. In the literature, many papers are devoted to the estimation of the position of Gaussian kernels, as...
A LOW-POWER SILICON-ON-INSULATOR PWM DISCRIMINATOR FOR BIOMEDICAL APPLICATIONS (2007)
Jader A. De Lima, Sidnei F. Silva, Adriano S. Cordeiro, Ro C. Araujo, Michel Verleysen
A CMOS/SOI circuit to decode PWM signals is presented as part of a body-implanted neurostimulator for visual prosthesis. Since encoded data is the sole input to the circuit, the decoding technique is...
A VLSI SYSTEM FOR NEURAL BAYESIAN AND LVQ CLASSIFICATION (2007)
Michel Verleysen, Jean-didier Legat, Jordi Madrenas Y, Jordi Dom Nguez Y, Universite Catholique De Louvain
Various types of neural networks may beusedinmulti-dimensional classi cation tasks � among them, Bayesian and LVQ algorithms are interesting respectively for their performances and their simplicity...
Optimal decision surfaces in LVQ1 classification of patterns (2007)
Michel Verleysen, Jean-didier Legat
Abstract. Kohonen's LVQ1 procedure is widely used for the classification of patterns in a multi-class distribution. This algorithm approximates the probability densities of vectors in each...
Eric De Bodt, Marie Cottrell, Michel Verleysen
Abstract. In a previous paper ([1], ESANN'97), we compared the Kohonen algorithm (SAM) to Simple Competitive Learning Algorithm (SCL) when the goal is to reconstruct an unknown density. We...
Map with On-Chip Learning (2007)
Damien Macq, Michel Verleysen, Paul Jespers, Jean-didier Legat
Abstract--Kohonen maps are self-organizing neural networks that classify and quantify n-dimensional data into a one- or two-dimensional array of neurons [1]-[2]. Most applications of Kohonen maps use...
APRACTICAL VIEW OF SUBOPTIMAL BAYESIAN CLASSIFICATION WITH RADIAL GAUSSIAN KERNELS (2007)
Jean-luc Voz, Michel Verleysen, Jean-didier Legat, Universite Catholique De Louvain
For pattern classi cation in a multi-dimensional space, the minimum misclassi cation rate is obtained by using the Bayes criterion. Kernel estimators or probabilistic neural networks provide a good...
C2.1 Feedforward models (2007)
Feedforward unsupervised models cover a wide range of neural networks with various applications. In this section, we discuss three widely used models. (i) Kohonen’s selforganizing map, also called...
Michel Verleysen, Marie Cottrell, Universit Catholique De Louvain, Pl. Des Doyens
Abstract. Besides their topological properties, Kohonen maps are often used for vector quantization only. These auto-organised networks are often compared to other standard and/or adaptive vector...
An optimized RBF network for approximation of functions (2007)
Michel Verleysen, Katerina Hlavackova
Abstract. RBF networks are widely used for the non-parametric estimation of real-valued multi-dimentional functions through a finite set of samples This paper describes a method to compute the...
John Aldo Lee, Michel Verleysen
Abstract. Recent papers have clearly shown the advantage of using the geodesic distance instead of the Euclidean one in methods performing non-linear dimensionality reduction by means of distance...
Fast Selection of Spectral Variables with B-Spline Compression (2007)
Rossi, Fabrice, François, Damien, Wertz, Vincent, Meurens, Marc, Verleysen, Michel
The large number of spectral variables in most data sets encountered in spectral chemometrics often renders the prediction of a dependent variable uneasy. The number of variables hopefully can be...
Resampling methods for parameter-free and robust feature selection with mutual information (2007)
François, Damien, Rossi, Fabrice, Wertz, Vincent, Verleysen, Michel
Combining the mutual information criterion with a forward feature selection strategy offers a good trade-off between optimality of the selected feature subset and computation time. However, it...
Representation of Functional Data in Neural Networks (2007)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Rossi, Fabrice, Lendasse, Amaury, François, Damien, Wertz, Vincent, Verleysen, Michel
Data from spectrophotometers form vectors of a large number of exploitable variables. Building quantitative models using these variables most often requires using a smaller set of variables than the...
Time series prediction competition: The CATS benchmark (2007)
Lendasse, Amaury, Oja, Erkki, Simula, Olli, Verleysen, Michel
Mixtures of Robust Probabilistic Principal Component Analyzers (2007)
Archambeau, Cedric, Delannay, Nicolas, Verleysen, Michel
Mixtures of probabilistic principal component analyzers model high-dimensional nonlinear data by combining local linear models. Each mixture component is specifically designed to extract the local...
Representation of Functional Data in Neural Networks (2007)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Representation of Functional Data in Neural Networks (2007)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Forecasting the CATS benchmark with the Double Vector Quantization method (2007)
Simon, Geoffroy, Lee, John, Cottrell, Marie, Verleysen, Michel
The Double Vector Quantization method, a long-term forecasting method based on the SOM algorithm, has been used to predict the 100 missing values of the CATS competition data set. An analysis of the...
Time Series Forecasting: Obtaining Long Term Trends with Self-Organizing Maps (2007)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Kohonen self-organisation maps are a well know classification tool, commonly used in a wide variety of problems, but with limited applications in time series forecasting context. In this paper, we...
On the use of self-organizing maps to accelerate vector quantization (2007)
De Bodt, Eric, Cottrell, Marie, Letrémy, Patrick, Verleysen, Michel
Self-organizing maps (SOM) are widely used for their topology preservation property: neighboring input vectors are quantified (or classified) either on the same location or on neighbor ones on a...
Statistical tools to assess the reliability of self-organizing maps (2007)
De Bodt, Eric, Cottrell, Marie, Verleysen, Michel
Results of neural network learning are always subject to some variability, due to the sensitivity to initial conditions, to convergence to local minima, and, sometimes more dramatically, to sampling...
Robust Bayesian Clustering (2007)
Archambeau, Cedric, Verleysen, Michel
A new variational Bayesian learning algorithm for Student-t mixture models is introduced. This algorithm leads to (i) robust density estimation, (ii) robust clustering and (iii) robust automatic...
Forecasting the CATS benchmark with the Double Vector Quantization method (2007)
Simon, Geoffroy, Lee, John, Cottrell, Marie, Verleysen, Michel
The Double Vector Quantization method, a long-term forecasting method based on the SOM algorithm, has been used to predict the 100 missing values of the CATS competition data set. An analysis of the...
Resampling methods for parameter-free and robust feature selection with mutual information (2007)
François, Damien, Rossi, Fabrice, Wertz, Vincent, Verleysen, Michel
Combining the mutual information criterion with a forward feature selection strategy offers a good trade-off between optimality of the selected feature subset and computation time. However, it...
Fast Selection of Spectral Variables with B-Spline Compression (2007)
Rossi, Fabrice, François, Damien, Wertz, Vincent, Meurens, Marc, Verleysen, Michel
The large number of spectral variables in most data sets encountered in spectral chemometrics often renders the prediction of a dependent variable uneasy. The number of variables hopefully can be...
Resampling methods for parameter-free and robust feature selection with mutual information (2007)
François, Damien, Rossi, Fabrice, Wertz, Vincent, Verleysen, Michel
Combining the mutual information criterion with a forward feature selection strategy offers a good trade-off between optimality of the selected feature subset and computation time. However, it...
Fast Selection of Spectral Variables with B-Spline Compression (2007)
Rossi, Fabrice, François, Damien, Wertz, Vincent, Meurens, Marc, Verleysen, Michel
The large number of spectral variables in most data sets encountered in spectral chemometrics often renders the prediction of a dependent variable uneasy. The number of variables hopefully can be...
DD-HDS: A method for visualization and exploration of high-dimensional data. (2007)
Lespinats, Sylvain, Verleysen, Michel, Giron, Alain, Fertil, Bernard
Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven high-dimensional...
DD-HDS: A method for visualization and exploration of high-dimensional data. (2007)
Lespinats, Sylvain, Verleysen, Michel, Giron, Alain, Fertil, Bernard
Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven high-dimensional...
Forecasting the CATS benchmark with the Double Vector Quantization method (2007)
Simon, Geoffroy, Lee, John, Cottrell, Marie, Verleysen, Michel
The Double Vector Quantization method, a long-term forecasting method based on the SOM algorithm, has been used to predict the 100 missing values of the CATS competition data set. An analysis of the...
Functional SOM for variable-length signal windows (2007)
De Decker, Arnaud, De Lannoy, Gael, Verleysen, Michel
Functional data, often sampled at high frequency, lead to high-dimensional vectors. The curse of dimensionality makes the latter difficult to handle with standard data analysis tools. Functional data...
Robust Bayesian clustering (2007)
Cédric Archambeau, Michel Verleysen
www.elsevier.com/locate/neunet A new variational Bayesian learning algorithm for Student-t mixture models is introduced. This algorithm leads to (i) robust density estimation, (ii) robust clustering...
The concentration of fractional distances (2007)
Damien François, Vincent Wertz, Michel Verleysen, Senior Member
Abstract—Nearest neighbor search and many other numerical data analysis tools most often rely on the use of the euclidean distance. When data are high dimensional, however, the euclidean distances...
Is the general form of Renyi’s entropy a contrast for source separation (2007)
Frédéric Vrins, Dinh-tuan Pham, Michel Verleysen
Abstract. Renyi’s entropy-based criterion has been proposed as an objective function for independent component analysis because of its relationship with Shannon’s entropy and its computational...
07131 Summary -- Similarity-based Clustering and its Application to Medicine and Biology (2007)
Biehl, Michael, Hammer, Barbara, Verleysen, Michel, Villmann, Thomas
This paper summarizes presentations, discussions, and results of the Dagstuhl seminar.
Biehl, Michael, Hammer, Barbara, Verleysen, Michel, Villmann, Thomas
From 25.03. to 30.03.2007, the Dagstuhl Seminar 07131 ``Similarity-based Clustering and its Application to Medicine and Biology'' was held in the International Conference and Research Center (IBFI),...
DD-HDS: A method for visualization and exploration of high-dimensional data. (2007)
Lespinats, Sylvain, Verleysen, Michel, Giron, Alain, Fertil, Bernard
Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven high-dimensional...
DD-HDS: A method for visualization and exploration of high-dimensional data. (2007)
Lespinats, Sylvain, Verleysen, Michel, Giron, Alain, Fertil, Bernard
Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven high-dimensional...
DD-HDS: A method for visualization and exploration of high-dimensional data. (2007)
Lespinats, Sylvain, Verleysen, Michel, Giron, Alain, Fertil, Bernard
Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven high-dimensional...
Fast Selection of Spectral Variables with B-Spline Compression (2007)
Rossi, Fabrice, François, Damien, Wertz, Vincent, Meurens, Marc, Verleysen, Michel
The large number of spectral variables in most data sets encountered in spectral chemometrics often renders the prediction of a dependent variable uneasy. The number of variables hopefully can be...
Resampling methods for parameter-free and robust feature selection with mutual information (2007)
François, Damien, Rossi, Fabrice, Wertz, Vincent, Verleysen, Michel
Combining the mutual information criterion with a forward feature selection strategy offers a good trade-off between optimality of the selected feature subset and computation time. However, it...
Forecasting the CATS benchmark with the Double Vector Quantization method (2007)
Simon, Geoffroy, Lee, John, Cottrell, Marie, Verleysen, Michel
The Double Vector Quantization method, a long-term forecasting method based on the SOM algorithm, has been used to predict the 100 missing values of the CATS competition data set. An analysis of the...
DD-HDS: A method for visualization and exploration of high-dimensional data. (2007)
Lespinats, Sylvain, Verleysen, Michel, Giron, Alain, Fertil, Bernard
Mapping high-dimensional data in a low-dimensional space, for example, for visualization, is a problem of increasingly major concern in data analysis. This paper presents data-driven high-dimensional...
Fast Selection of Spectral Variables with B-Spline Compression (2007)
Rossi, Fabrice, François, Damien, Wertz, Vincent, Meurens, Marc, Verleysen, Michel
The large number of spectral variables in most data sets encountered in spectral chemometrics often renders the prediction of a dependent variable uneasy. The number of variables hopefully can be...
Resampling methods for parameter-free and robust feature selection with mutual information (2007)
François, Damien, Rossi, Fabrice, Wertz, Vincent, Verleysen, Michel
Combining the mutual information criterion with a forward feature selection strategy offers a good trade-off between optimality of the selected feature subset and computation time. However, it...
Forecasting the CATS benchmark with the Double Vector Quantization method (2007)
Simon, Geoffroy, Lee, John, Cottrell, Marie, Verleysen, Michel
The Double Vector Quantization method, a long-term forecasting method based on the SOM algorithm, has been used to predict the 100 missing values of the CATS competition data set. An analysis of the...
M.: Mixtures of robust probabilistic principal component analysers (2007)
Cédric Archambeau, Nicolas Delannay, Michel Verleysen
Abstract. Discovering low-dimensional (nonlinear) manifolds is an important problem in Machine Learning. In many applications, the data are in a high dimensional space. This can be problematic in...
Robust Bayesian clustering (2007)
Cédric Archambeau, Michel Verleysen
A new variational Bayesian learning algorithm for Student-t mixture models is introduced. This algorithm leads to (i) robust density estimation, (ii) robust clustering and (iii) robust automatic...
Advances in Self Organising Maps (2006)
Cottrell, Marie, Verleysen, Michel
The Self-Organizing Map (SOM) with its related extensions is the most popular artificial neural algorithm for use in unsupervised learning, clustering, classification and data visualization. Over...
Automatic Adjustment of Discriminant Adaptive Nearest Neighbor (2006)
Delannay, Nicolas, Archambeau, Cedric, Verleysen, Michel
K-Nearest Neighbors relies on the definition of a global metric. In contrast, Discriminant Adaptive Nearest Neighbor (DANN) computes a different metric at each query point based on a local Linear...
Robust Probabilistic Projections (2006)
Archambeau, Cedric, Delannay, Nicolas, Verleysen, Michel
Principal components and canonical correlations are at the root of many exploratory data mining techniques and provide standard pre-processing tools in machine learning. Lately, probabilistic...
Determination of the Mahalanobis matrix using nonparametric noise estimations (2006)
Lendasse, Amaury, Corona, Francesco, Hao, Jin, Reyhani, Nima, Verleysen, Michel
In this paper, the problem of an optimal transformation of the input space for function approximation problems is ddressed. The transformation is defined determining the Mahalanobis matrix that...
Robust Probabilistic Projections (2006)
Archambeau, Cedric, Verleysen, Michel
Principal components and canonical correlations are at the root of many exploratory data mining techniques and provide standard pre-processing tools in machine learning. Lately, probabilistic...
Advances in Self Organising Maps (2006)
Cottrell, Marie, Verleysen, Michel
The Self-Organizing Map (SOM) with its related extensions is the most popular artificial neural algorithm for use in unsupervised learning, clustering, classification and data visualization. Over...
Rossi, Fabrice, Lendasse, Amaury, François, Damien, Wertz, Vincent, Verleysen, Michel
Data from spectrophotometers form vectors of a large number of exploitable variables. Building quantitative models using these variables most often requires using a smaller set of variables than the...
Rossi, Fabrice, Lendasse, Amaury, François, Damien, Wertz, Vincent, Verleysen, Michel
Data from spectrophotometers form vectors of a large number of exploitable variables. Building quantitative models using these variables most often requires using a smaller set of variables than the...
Advances in Self Organising Maps (2006)
Cottrell, Marie, Verleysen, Michel
The Self-Organizing Map (SOM) with its related extensions is the most popular artificial neural algorithm for use in unsupervised learning, clustering, classification and data visualization. Over...
Robust probabilistic projections (2006)
Cédric Archambeau, Nicolas Delannay, Michel Verleysen
Principal components and canonical correlations are at the root of many exploratory data mining techniques and provide standard pre-processing tools in machine learning. Lately, probabilistic...
Non-orthogonal support width ica (2006)
John A. Lee, Frédéric Vrins, Michel Verleysen
Abstract. Independent Component Analysis (ICA) is a powerful tool with applications in many areas of blind signal processing; however, its key assumption, i.e. the statistical independence of the...
Minimum support ICA using order statistics. Part I: Quasi-range based support estimation (2006)
Frédéric Vrins, Michel Verleysen
Abstract. Linear instantaneous independent component analysis (ICA) is a well-known problem, for which efficient algorithms like FastICA and JADE have been developed. Nevertheless, the development of...
Minimum support ICA using order statistics. Part I: Quasi-range based support estimation (2006)
Frédéric Vrins, Michel Verleysen
Abstract. The minimum support ICA algorithms currently use the extreme statistics difference (also called the statistical range) for support width estimation. In this paper, we extend this method by...
Robust probabilistic projections (2006)
Cédric Archambeau, Nicolas Delannay, Michel Verleysen
Principal components and canonical correlations are at the root of many exploratory data mining techniques and provide standard pre-processing tools in machine learning. Lately, probabilistic...
Non-orthogonal support width ica (2006)
John A. Lee, Frédéric Vrins, Michel Verleysen
Abstract. Independent Component Analysis (ICA) is a powerful tool with applications in many areas of blind signal processing; however, its key assumption, i.e. the statistical independence of the...
Advances in Self Organising Maps (2006)
Cottrell, Marie, Verleysen, Michel
The Self-Organizing Map (SOM) with its related extensions is the most popular artificial neural algorithm for use in unsupervised learning, clustering, classification and data visualization. Over...
Rossi, Fabrice, Lendasse, Amaury, François, Damien, Wertz, Vincent, Verleysen, Michel
Data from spectrophotometers form vectors of a large number of exploitable variables. Building quantitative models using these variables most often requires using a smaller set of variables than the...
Advances in Self Organising Maps (2006)
Cottrell, Marie, Verleysen, Michel
The Self-Organizing Map (SOM) with its related extensions is the most popular artificial neural algorithm for use in unsupervised learning, clustering, classification and data visualization. Over...
Rossi, Fabrice, Lendasse, Amaury, François, Damien, Wertz, Vincent, Verleysen, Michel
Data from spectrophotometers form vectors of a large number of exploitable variables. Building quantitative models using these variables most often requires using a smaller set of variables than the...
LS-SVM hyperparameter selection with a nonparametric noise estimator (2005)
Lendasse, Amaury, Ji, Yongnan, Reyhani, Nima, Verleysen, Michel
This paper presents a new method for the selection of the two hyperparameters of Least Squares Support Vector Machine (LS-SVM) approximators with Gaussian Kernels. The two hyperparameters are the...
Time series forecasting: obtaining long term trends with self-organizing maps (2005)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Kohonen self-organisation maps are a well know classification tool, commonly used in a wide variety of problems, but with limited applications in time series forecasting context. In this paper, we...
Vector quantization: a weighted version for time-series forecasting (2005)
Lendasse, Amaury, Francois, Damien, Wertz, Vincent, Verleysen, Michel
Nonlinear time-series prediction offers potential performance increases compared to linear models. Nevertheless, the enhanced complexity and computation time often prohibits an efficient use of...
Pruned lazy learning models for time series prediction (2005)
Sorjamaa, Antti, Lendasse, Amaury, Verleysen, Michel
This paper presents two improvements of Lazy Learning. Both methods include input selection and are applied to the long-term prediction of time series. The first method is based on an iterative...
Representation of Functional Data in Neural Networks (2005)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Fast bootstrap methodology for model selection (2005)
Lendasse, Amaury, Simon, Geoffroy, Wertz, Vincent, Verleysen, Michel
Using resampling methods like cross-validation and bootstrap is a necessity in neural network design, for solving the problem of model structure selection. The bootstrap is a powerful method offering...
Manifold Constrained Finite Gaussian Mixtures (2005)
Archambeau, Cedric, Verleysen, Michel
In many practical applications, the data is organized along a manifold of lower dimension than the dimension of the embedding space. This additional information can be used when learning the model...
Local Vector-based Models for Sense Discrimination (2005)
De Marneffe, Marie-Catherine, Archambeau, Cedric, Dupont, Pierre, Verleysen, Michel
Word sense discrimination aims at automatically determining which instances of an ambiguous word share the same sense. A fully unsupervised technique based on a high dimensional vector representation...
Manifold Constrained Variational Mixtures (2005)
Archambeau, Cedric, Verleysen, Michel
In many data mining applications, the data manifold is of lower dimension than the dimension of the input space. In this paper, it is proposed to take advantage of this additional information in the...
Time Series Forecasting: Obtaining Long Term Trends with Self-Organizing Maps (2005)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Kohonen self-organisation maps are a well know classification tool, commonly used in a wide variety of problems, but with limited applications in time series forecasting context. In this paper, we...
Representation of Functional Data in Neural Networks (2005)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Representation of Functional Data in Neural Networks (2005)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Time Series Forecasting: Obtaining Long Term Trends with Self-Organizing Maps (2005)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Kohonen self-organisation maps are a well know classification tool, commonly used in a wide variety of problems, but with limited applications in time series forecasting context. In this paper, we...
Filtering-Free Blind Separation of Correlated Images (2005)
Frederic Vrins, John A. Lee, Michel Verleysen
When using ICA for image separation, a well-known problem is that most often a large correlation exists between the sources.
SignalProcu.I-C 85 (2005) 1029--1044 (2005)
On The Entropy, Fre Déric Vrins, Michel Verleysen
The marginal entropy hZ of a weighted sum of two variables Z aX bY ; expressed as a func.C' of its weights, is a usualcua func`IE for blind sourc separation (BSS), and moreprecECkk for...
Information Theoretic versus Cumulant-Based Contrasts for Multimodal Source Separation (2005)
Frederic Vrins, Michel Verleysen, Senior Member
Recently, several authors have emphasized the existence of spurious maxima in usual contrast functions for source separation (e.g., the likelihood and the mutual information) when several sources...
Nonlinear Dimensionality Reduction of Data Manifolds With Essential Loops (2005)
John Aldo Lee, Michel Verleysen, Communicated S. Fiori
Numerous methods or algorithms have been designed to solve the problem of nonlinear dimensionality reduction (NLDR). However, very few among them are able to embed efficiently `circular'...
Clustering Using a Random Walk Based Distance Measure (2005)
Luh Yen, Denis Vanvyve, Fabien Wouters, François Fouss, Michel Verleysen, Marco Saerens
This work proposes a simple way to improve a clustering algorithm.
The Growth Trajectories Of Start-Up Firms: (2005)
An Exploratory Study, Benoît Gailly, Damien François, Mahamadou Biga Diambeidou, Michel Verleysen
this paper discusses the existence of different growth trajectories of start-ups. Using financial data from all firms created from 1992 to 2002 in Belgium (N=152064), we identified all those that had...
A Simple ICA Algorithm for Non-Differentiable Contrasts (2005)
John A. Lee, Frederic Vrins, Michel Verleysen
A general-purpose algorithm is proposed for Independent Component Analysis. This algorithm is specifically designed in order to handle non-differentiable contrast functions. Sources are extracted one...
Can We Always Trust Entropy Minima In The Ica Context ? (2005)
Frederic Vrins John, John A. Lee, Michel Verleysen
Marginal entropy can be used as cost function for blind source separation (BSS). Recently, some authors have experimentally shown that such information-theoretic cost function may have spurious...
The curse of dimensionality in data mining and time series prediction (2005)
Michel Verleysen, Damien François
www.ucl.ac.be/mlg Abstract. Modern data analysis tools have to work on high-dimensional data, whose components are not independently distributed. High-dimensional spaces show surprising,...
SWM : a Class of Convex Contrasts for Source Separation (2005)
Frédéric Vrins, Michel Verleysen, Place Du Levant
We derive a class of contrasts for blind source separation (BSS) to separate bounded sources (or more generally, finite sources), based on support width measures (SWM) of the marginal output...
Pruned Lazy Learning Models for Time Series Prediction (2005)
Antti Sorjamaa, Amaury Lendasse, Michel Verleysen
Abstract. This paper presents two improvements of Lazy Learning. Both methods include input selection and are applied to long-term prediction of time series. First method is based on an iterative...
On the need of unfolding preprocessing for time series clustering (2005)
Geoffroy Simon, John A. Lee, Michel Verleysen
Abstract- Clustering methods are commonly used on time series, either as a preprocessing for other methods or for themselves. This paper illustrates the problem of clustering applied on regressor...
Filtering-free Blind Separation of Correlated Images (2005)
Frédéric Vrins, John A. Lee, Michel Verleysen
Abstract. When using ICA for image separation, a well-known problem is that most often a large correlation exists between the sources. Because of this dependence, there is no more guarantee that the...
Representation of Functional Data in Neural Networks (2005)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Time Series Forecasting: Obtaining Long Term Trends with Self-Organizing Maps (2005)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Kohonen self-organisation maps are a well know classification tool, commonly used in a wide variety of problems, but with limited applications in time series forecasting context. In this paper, we...
Representation of Functional Data in Neural Networks (2005)
Rossi, Fabrice, Delannay, Nicolas, Conan-Guez, Brieuc, Verleysen, Michel
Functional Data Analysis (FDA) is an extension of traditional data analysis to functional data, for example spectra, temporal series, spatio-temporal images, gesture recognition data, etc. Functional...
Time Series Forecasting: Obtaining Long Term Trends with Self-Organizing Maps (2005)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Kohonen self-organisation maps are a well know classification tool, commonly used in a wide variety of problems, but with limited applications in time series forecasting context. In this paper, we...
Rossi, Fabrice, Lendasse, Amaury, Francois, Damien, Wertz, Vincent, Verleysen, Michel
Data from spectrophotometers form vectors of a large number of exploitable variables. Building quantitative models using these variables most often requires using a smaller set of variables than the...
Vrins, Frederic, Archambeau, Cedric, Verleysen, Michel
The source separation problem is usually solved through a gradient descent on a cost function C. However, C may have local minima that are irrelevant from the source separation point of view in...
Supervised Nonparametric Information Theoretic classification (2004)
Archambeau, Cedric, Butz, Tortzen, Popovici, Vlad, Verleysen, Michel, Thiran, Jean-Philippe
In this paper, supervised nonparametric information theoretic classification (ITC) is introduced. Its principle relies on the likelihood of a data sample of transmitting its class label to data...
Towards a Local Separation Performances Estimator Using Common ICA Contrast Functions ? (2004)
Vrins, Frederic, Archambeau, Cedric, Verleysen, Michel
In most ICA algorithms, the separation performances are estimated through the evaluation of a contrast function Phi, used in the update rule of elements of the unmixing matrix. In particular...
Flexible and Robust Bayesian Classification by Finite Mixture Models (2004)
Archambeau, Cedric, Vrins, Frederic, Verleysen, Michel
The regularized Mahalanobis distance is proposed in the framework of Finite mixture models to avoid commonly faced numerical difficulties encountered with EM. Its principle is applied to Gaussian and...
Archambeau, Cedric, Delbeke, Jean, Veraart, Claude, Verleysen, Michel
Within the framework of the OPTIVIP project, an optic nerve based visual prosthesis is developed in order to restore partial vision to the blind. One of the main challenges is to understand, decode...
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
The Kohonen self-organization map is usually considered as a classification or clustering tool, with only a few applications in time series prediction. In this paper, a particular time series...
On the use of self-organizing maps to accelerate vector quantization (2004)
De Bodt, Eric, Cottrell, Marie, Letrémy, Patrick, Verleysen, Michel
Self-organizing maps (SOM) are widely used for their topology preservation property: neighboring input vectors are quantified (or classified) either on the same location or on neighbor ones on a...
On the use of self-organizing maps to accelerate vector quantization (2004)
De Bodt, Eric, Cottrell, Marie, Letrémy, Patrick, Verleysen, Michel
Self-organizing maps (SOM) are widely used for their topology preservation property: neighboring input vectors are quantified (or classified) either on the same location or on neighbor ones on a...
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
The Kohonen self-organization map is usually considered as a classification or clustering tool, with only a few applications in time series prediction. In this paper, a particular time series...
M.: Towards a local separation performances estimator using common ica contrast functions (2004)
Frédéric Vrins, Cédric Archambeau, Michel Verleysen
Abstract. In most ICA algorithms, the separation performances are estimated through the evaluation of a contrast function Φ, used in the update rule of elements of the unmixing matrix. In particular...
Nonlinear projection with curvilinear distances: Isomap versus curvilinear distance analysis (2004)
John Aldo Lee, Amaury Lendasse, Michel Verleysen
Didixdx reductid technidxd arewidCC used for theanalysi andviPHx#EdqPTP of complex sets of data.Thi paper compares two recently publilyd methods fornonliTEH projectiqP Isomap andCurvikdqCE DidCurv...
Flexible and Robust Bayesian Classification by Finite Mixture Models (2004)
Cedric Archambeau, Federic Vrins, Michel Verleysen
The regularized Mahalanobis distance is proposed in the framework of finite mixture models to avoid commonly faced numerical difficulties encountered with EM. Its principle is applied to Gaussian and...
Double Quantization Forecasting Method for Filling Missing Data in the CATS Time Series (2004)
Geoffroy Simon John, John A. Lee, Michel Verleysen
The double vector quantization forecasting method based on Kohonen self-organizing maps is applied to predict the missing values of the CATS Competition data set. As one of the features of the method...
Frederic Vrins, Christian Jutten, Michel Verleysen
Recently, non-invasive techniques to measure the fetal electrocardiogram (FECG) signal have given very promising results. However, the important question of the number and the location of the...
M.: Towards a local separation performances estimator using common ica contrast functions (2004)
Frédéric Vrins, Cédric Archambeau, Michel Verleysen
Abstract. In most ICA algorithms, the separation performances are estimated through the evaluation of a contrast function Φ, used in the update rule of elements of the unmixing matrix. In particular...
On the use of self-organizing maps to accelerate vector quantization (2004)
De Bodt, Eric, Cottrell, Marie, Letrémy, Patrick, Verleysen, Michel
Self-organizing maps (SOM) are widely used for their topology preservation property: neighboring input vectors are quantified (or classified) either on the same location or on neighbor ones on a...
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
The Kohonen self-organization map is usually considered as a classification or clustering tool, with only a few applications in time series prediction. In this paper, a particular time series...
On the use of self-organizing maps to accelerate vector quantization (2004)
De Bodt, Eric, Cottrell, Marie, Letrémy, Patrick, Verleysen, Michel
Self-organizing maps (SOM) are widely used for their topology preservation property: neighboring input vectors are quantified (or classified) either on the same location or on neighbor ones on a...
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
The Kohonen self-organization map is usually considered as a classification or clustering tool, with only a few applications in time series prediction. In this paper, a particular time series...
Classification of Visual Sensations Generated Electrically in the Visual Field of the Blind (2003)
Archambeau, Cedric, Delbeke, Jean, Verleysen, Michel
Within the framework of the OPTIVIP project, an optic nerve based visual prosthesis is being developed in order to restore partial vision to the blind. In this paper, we concentrate on the...
On Convergence Problems of the EM Algorithm for Finite Gaussian Mixtures (2003)
Archambeau, Cedric, Lee, John A, Verleysen, Michel
Efficient probability density function estimation is of primary interest in statistics. A popular approach for achieving this is the use of finite Gaussian mixture models. Based on the...
Locally Linear Embedding versus Isotop (2003)
Lee, John A, Archambeau, Cedric, Verleysen, Michel
Recently, a new method intended to realize conformal map- pings has been published. Called Locally Linear Embedding (LLE), this method can map high-dimensional data lying on a manifold to a represen-...
Archambeau, Cedric, Verleysen, Michel
Flexible and reliable probability density estimation is fundamental in unsupervised learning and classification. Finite Gaussian mixture models are commonly used to serve this purpose. However, they...
Dualibe, Carlos, Verleysen, Michel, Jespers, Paul G.A.
Contenido: Lógica difusa y sistemas difusos; Bloques de construcción análogos básicos; Lógica difusa programable de señales mixtas; Controladores; Análisis de señales tiempo-dominio usando la...
Dualibe, Carlos, Verleysen, Michel, Jespers, Paul G.A.
1-4020-7359-3
Double SOM for long-term time series prediction (2003)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Many time series forecasting problems require the estimation of possibly inaccurate, but long¬term, trends, rather than accurate short-term prediction. In this paper, a double use of the...
Double SOM for long-term time series prediction (2003)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Many time series forecasting problems require the estimation of possibly inaccurate, but long¬term, trends, rather than accurate short-term prediction. In this paper, a double use of the...
Amaury Lendasse, Michel Verleysen
Abstract. This paper compares several model selection methods, based on experimental estimates of their generalization errors. Experiments in the context of nonlinear time series prediction by...
On convergence problems of the em algorithm for finite gaussian mixtures (2003)
Cédric Archambeau, John A. Lee, Michel Verleysen
Abstract. Efficient probability density function estimation is of primary interest in statistics. A popular approach for achieving this is the use of finite Gaussian mixture models. Based on the...
Locally Linear Embedding versus Isotop (2003)
John Aldo Lee, Cédric Archambeau, Michel Verleysen
Recently, a new method intended to realize conformal mappings has been published. Called Locally Linear Embedding (LLE), this method can map high-dimensional data lying on a manifold to a...
On the kernel widths in radial-basis function networks, Neural Process (2003)
Nabil Benoudjit, Michel Verleysen
Abstract. RBFN (Radial-Basis Function Networks) represent an attractive alternative to other neural network models. Their learning is usually split into an unsupervised part, where center and widths...
Cédric Archambeau, Michel Verleysen
Flexible and reliable probability density estimation is fundamental in unsupervised learning and classification. Finite Gaussian mixture models are commonly used to serve this purpose. However, they...
Simon Dablemont, Geoffroy Simon, Amaury Lendasse, Alain Ruttiens, Francois Blayo, Michel Verleysen
A general method for time series forecasting is presented. Based on the splitting of the past dynamics into clusters, local models are built to capture the possible evolution of the series given the...
Classification of Visual Sensations Generated Electrically in the Visual Field of the Blind (2003)
Cedric Archambeau, Jean Delbeke, Michel Verleysen
Within the framework of the OPTIVIP project, an optic nerve based visual prosthesis is being developed in order to restore partial vision to the blind. In this paper, we concentrate on the...
Double SOM for long-term time series prediction (2003)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Many time series forecasting problems require the estimation of possibly inaccurate, but long¬term, trends, rather than accurate short-term prediction. In this paper, a double use of the...
Double SOM for long-term time series prediction (2003)
Simon, Geoffroy, Lendasse, Amaury, Cottrell, Marie, Fort, Jean-Claude, Verleysen, Michel
Many time series forecasting problems require the estimation of possibly inaccurate, but long¬term, trends, rather than accurate short-term prediction. In this paper, a double use of the...
Width Optimization of the Gaussian kernels in Radial Basis Function Networks (2002)
Benoudjit, Nabil, Archambeau, Cedric, Lendasse, Amaury, Lee, John A, Verleysen, Michel
Radial basis function networks are usually trained according to a three-stage procedure. In the literature, many papers are devoted to the estimation of the position of Gaussian kernels, as well as...
Statistical tools to assess the reliability of self-organizing maps (2002)
De Bodt, Eric, Cottrell, Marie, Verleysen, Michel
Results of neural network learning are always subject to some variability, due to the sensitivity to initial conditions, to convergence to local minima, and, sometimes more dramatically, to sampling...
Statistical tools to assess the reliability of self-organizing maps (2002)
De Bodt, Eric, Cottrell, Marie, Verleysen, Michel
Results of neural network learning are always subject to some variability, due to the sensitivity to initial conditions, to convergence to local minima, and, sometimes more dramatically, to sampling...
Nonlinear Projection with the Isotop Method (2002)
Abstract. Isotop is a new neural method for nonlinear projection of high-dimensional data. Isotop builds the mapping between the data space and a projection space by means of topology preservation....
Curvilinear distance analysis versus isomap (2002)
John Aldo Lee, Amaury Lendasse, Michel Verleysen
Abstract. Dimension reduction techniques are widely used for the analysis and visualization of complex sets of data. This paper compares two nonlinear projection methods: Isomap and Curvilinear...
A CMOS/SOI Single-input PWM Discriminator for Low-voltage Body-implanted Applications (2002)
Jader A. De Lima, Sidnei F. Silva, Adriano S. Cordeiro, Michel Verleysen
A CMOS/SOI circuit to decode Pulse-Width Modulation (PWM) signals is presented as part of a body-implanted neurostimulator for visual prosthesis. Since encoded data is the sole input to the circuit,...
Statistical tools to assess the reliability of self-organizing maps (2002)
De Bodt, Eric, Cottrell, Marie, Verleysen, Michel
Results of neural network learning are always subject to some variability, due to the sensitivity to initial conditions, to convergence to local minima, and, sometimes more dramatically, to sampling...
Statistical tools to assess the reliability of self-organizing maps (2002)
De Bodt, Eric, Cottrell, Marie, Verleysen, Michel
Results of neural network learning are always subject to some variability, due to the sensitivity to initial conditions, to convergence to local minima, and, sometimes more dramatically, to sampling...
Phosphene Evaluation in a Visual Prosthesis with Artificial Neural Networks (2001)
Archambeau, Cedric, Lendasse, Amaury, Trullemans, Charles, Veraart, Claude, Delbeke, Jean, Verleysen, Michel
The electrical stimulation of the optic nerve is investigated, as an approach to the development of a microsystems-based visual prosthesis. Non-linear prediction models, i.e. artificial neural...
Amaury Lendasse, John Lee, Éric De Bodt, Vincent Wertz, Michel Verleysen, Member Ieee, ...
Abstract. – Prediction of financial time series using artificial neural networks has been the subject of many publications, even if the predictability of financial series remains a subject of...
Recursive learning rules for soms (2001)
John A. Lee, Nicolas Donckers, Michel Verleysen
Summary. Three extensions of the traditional learning rule for Self-Organizing Maps are presented. They are based on geometrical considerations and explore various possibilities regarding the norm...
Cédric Archambeau, Charles Trullemans, Michel Verleysen
ABSTRACT: The electrical stimulation of the optic nerve is investigated, as an approach to the development of a microsystems-based visual prosthesis. Non-linear prediction models, i.e. artificial...
Cédric Archambeau, Charles Trullemans, Michel Verleysen
ABSTRACT: The electrical stimulation of the optic nerve is investigated, as an approach to the development of a microsystems-based visual prosthesis. Non-linear prediction models, i.e. artificial...
A 5.26 Mflips programmable analogue fuzzy logic controller in a standard CMOS 2.4µ Technology (2000)
Carlos Dualibe, Paul Jespers, Michel Verleysen
A complete digitally- programmable analogue Fuzzy Logic Controller (FLC) is presented. The design of some new functional blocks and the improvement of others aim towards speed optimisation with a...
Design of complementary low-power CMOS architectures for loser-take-all and winner-take-all (1999)
Nicolas Donckers, Carlos Dualibe, Michel Verleysen
A novel architecture for winner-take-all (WTA) and looser-take-all (LTA) circuits is proposed. As compared with other realisations, the LTA does not require input subtraction from a reference, which...
Michel Verleysen, Bruno Sirletti, Andre M. Vandemeulebroecke, Andpaul G. A. Jespers
Abstract —Neural networks used as content-addressable memories show unequaled retrieval and speed capabilities in problems srreh as vision and pattern recognition. We propose a new implementation...
Image Compression by Self-Organized Kohonen Map (1998)
Christophe Amerijckx, Associate Member, Michel Verleysen, Jean-didier Legat
Abstract—This paper presents a compression scheme for digital still images, by using the Kohonen’s neural network algorithm, not only for its vector quantization feature, but also for its...
Forecasting time-series by Kohonen classification (1998)
Amaury Lendasse, Michel Verleysen, Eric De Bodt, Marie Cottrell
Abstract. In this paper, we propose a generic non-linear approach for time series forecasting. The main feature of this approach is the use of a simple statistical forecasting in small regions of an...
Bounds on the Degree of High Order Binary Perceptrons (1996)
Mayoraz, Eddy, Blayo, François, Verleysen, Michel
High order perceptrons are often used in order to reduce the size of neural networks. The complexity of the architecture of a usual multilayer network is then turned into the complexity of the...
Bounds on the Degree of High Order Binary Perceptrons (1996)
Mayoraz, Eddy, Blayo, François, Verleysen, Michel
High order perceptrons are often used in order to reduce the size of neural networks. The complexity of the architecture of a usual multilayer network is then turned into the complexity of the...
A statistical neural network for high-dimensional vector classification (1995)
Michel Verleysen, Jean�didier Legat
verleysen�dice.ucl.ac.be The minimum number of misclassi�cations in a multi�class classi�er is reached when the borders between classes are set according to the Bayes criterion. Unfortunately...
Estimation of performance bounds in supervised classification (1994)
Pierre Comon, Jean-luc Voz, Michel Verleysen
Abstract. The Bayes theory gives the ultimate performances that can be reached in a classification problem. We present in this paper a method that allows to estimate these performance bounds given...
An analog processor architecture for neural network classifier (1994)
Michel Verleysen, Philippe Thissen, Lean-Luc Voz, Jordi Madrenas
Many neural-like algorithms currently under study support classification tasks. Several of these algorithms base their functionality on LVQ-like procedures to find locations of centroids in the data...
A highstorage capacity content-addressable memory and its learning algorithm (1989)
Michel Verleysen, Bruno Sirletti, Paul G. A. Jespers
Abstrud-Hopfield’s neural networks show retrieval and speed capabili-ties that make them good candidates for content-addressable memories (CAM’s) in problems such as pattern recognition and...
Learning High-Dimensional Data
Michel Verleysen Universit, Michel Verleysen
Observations from real-world problems are often highdimensional vectors, i.e. made up of many variables. Learning methods, including artificial neural networks, often have difficulties to handle a...