Gilles Fleury
Supélec
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Gilles Fleury.
IEEE Transactions on Power Systems | 2005
Fouad Abou Chacra; Patrick Bastard; Gilles Fleury; Régine Clavreul
This paper presents a methodology to evaluate the impact of energy storage specific costs on net present value (NPV) of energy storage installations in distribution substations. This work is based on multiple objective optimization. Specific cost effects on economic performance of energy storage technologies are evaluated for an HV/MV substation. For each studied technology, sets of optimal economic operation strategies and capacities of the storage device are determined. Finally, this approach using nondominated sorting genetic algorithms (NSGAs) optimization technique achieves a practical method for load management, including storage in power distribution systems and produces diagrams defining relationship between specific costs and storage project payback.
EURASIP Journal on Advances in Signal Processing | 2004
Alfred O. Hero; Gilles Fleury; Alan J. Mears; Anand Swaroop
This paper introduces a statistical methodology for the identification of differentially expressed genes in DNA microarray experiments based on multiple criteria. These criteria are false discovery rate (FDR), variance-normalized differential expression levels (paired statistics), and minimum acceptable difference (MAD). The methodology also provides a set of simultaneous FDR confidence intervals on the true expression differences. The analysis can be implemented as a two-stage algorithm in which there is an initial screen that controls only FDR, which is then followed by a second screen which controls both FDR and MAD. It can also be implemented by computing and thresholding the set of FDR values for each gene that satisfies the MAD criterion. We illustrate the procedure to identify differentially expressed genes from a wild type versus knockout comparison of microarray data.
Physics in Medicine and Biology | 2010
A El Habachi; Emmanuelle Conil; Abdelhamid Hadjem; Emmanuel Vazquez; M.F. Wong; A. Gati; Gilles Fleury; Joe Wiart
In this paper, we propose identification of the morphological factors that may impact the whole-body averaged specific absorption rate (WBSAR). This study is conducted for the case of exposure to a front plane wave at a 2100 MHz frequency carrier. This study is based on the development of different regression models for estimating the WBSAR as a function of morphological factors. For this purpose, a database of 12 anatomical human models (phantoms) has been considered. Also, 18 supplementary phantoms obtained using the morphing technique were generated to build the required relation. This paper presents three models based on external morphological factors such as the body surface area, the body mass index or the body mass. These models show good results in estimating the WBSAR (<10%) for families obtained by the morphing technique, but these are still less accurate (30%) when applied to different original phantoms. This study stresses the importance of the internal morphological factors such as muscle and fat proportions in characterization of the WBSAR. The regression models are then improved using internal morphological factors with an estimation error of approximately 10% on the WBSAR. Finally, this study is suitable for establishing the statistical distribution of the WBSAR for a given population characterized by its morphology.
IEEE Transactions on Signal Processing | 1997
Sina Mirsaidi; Gilles Fleury; Jacques Oksman
This paper presents a new recursive algorithm for the time domain reconstruction and spectral estimation of uniformly sampled signals with missing observations. An autoregressive (AR) modeling approach is adopted. The AR parameters are estimated by optimizing a mean-square error criterion. The optimum is reached by means of a gradient method adapted to the nonperiodic sampling. The time-domain reconstruction is based on the signal prediction using the estimated model. The power spectral density is obtained using the estimated AR parameters. The development of the different steps of the algorithm is discussed in detail, and several examples are presented to demonstrate the practical results that can be obtained. The spectral estimates are compared with those obtained by known AR estimators applied to the same signals sampled periodically. We note that this algorithm can also be used in the case of nonstationary signals.
IEEE Transactions on Instrumentation and Measurement | 2006
J.I. De la Rosa; Gilles Fleury
In this paper, a new approach for the statistical characterization of a measurand is presented. A description of how different bootstrap techniques can be applied in practice to estimate successfully a measurand probability density function (pdf) is given. When the direct observation of a quantity of interest is practically impossible such as in nondestructive testing, it is necessary to estimate such quantity, which is also called measurand. The statistical characterization of any estimator is important, because all the uncertainty features can be accessible to qualify such estimator. On the other hand, most of the time, the large-scale repetition of an experiment is not economically feasible, so that the Monte Carlo methods cannot be used directly for uncertainty characterization. Bootstrap methods have proved to be a potentially useful alternative. Moreover, a biased bootstrap recent technique, with which robust parameter estimates are obtained, is used. This technique is extended to be used in measurand estimation. An extended nested bootstrap improvement for the measurand pdf estimation is also presented. These techniques are applied to a realistic multidimensional measurand-estimation problem of groove dimensioning using remote field eddy current inspection. Measurand uncertainty characterization using the bootstrap techniques generally gives an accurate pdf estimation
IEEE Signal Processing Letters | 2011
Arnau Tibau Puig; Ami Wiesel; Gilles Fleury; Alfred O. Hero
The scalar shrinkage-thresholding operator is a key ingredient in variable selection algorithms arising in wavelet denoising, JPEG2000 image compression and predictive analysis of gene microarray data. In these applications, the decision to select a scalar variable is given as the solution to a scalar sparsity penalized quadratic optimization. In some other applications, one seeks to select multidimensional variables. In this work, we present a natural multidimensional extension of the scalar shrinkage thresholding operator. Similarly to the scalar case, the threshold is determined by the minimization of a convex quadratic form plus an Euclidean norm penalty, however, here the optimization is performed over a domain of dimension N ≥ 1. The solution to this convex optimization problem is called the multidimensional shrinkage threshold operator (MSTO). The MSTO reduces to the scalar case in the special case of N=1. In the general case of N >; 1 the optimal MSTO shrinkage can be found through a simple convex line search. We give an efficient algorithm for solving this line search and show that our method to evaluate the MSTO outperforms other state-of-the art optimization approaches. We present several illustrative applications of the MSTO in the context of Group LASSO penalized estimation.
signal processing systems | 2004
Alfred O. Hero; Gilles Fleury
The massive scale and variability of microarray gene data creates new and challenging problems of signal extraction, gene clustering, and data mining, especially for temporal gene profiles. Many data mining methods for finding interesting gene expression patterns are based on thresholding single discriminants, e.g. the ratio of between-class to within-class variation or correlation to a template. Here a different approach is introduced for extracting information from gene microarrays. The approach is based on multiple objective optimization and we call it Pareto front analysis (PFA). This method establishes a ranking of genes according to estimated probabilities that each gene is Pareto-optimal, i.e., that it lies on the Pareto front of the multiple objective scattergram. Both a model-driven Bayesian Pareto method and a data-driven non-parametric Pareto method, based on rank-order statistics, are presented. The methods are illustrated for two gene microarray experiments.
international conference on acoustics, speech, and signal processing | 2003
Abd-Krim Seghouane; Maïza Bekara; Gilles Fleury
The Kullback information criterion (KIC) is a recently developed tool for statistical model selection (Cavanaugh, J.E., Statistics and Probability Letters, vol.42, p.333-43, 1999). KIC serves as an asymptotically unbiased estimator of a variant of the Kullback symmetric divergence, known also as J-divergence. A bias correction of the Kullback symmetric information criterion is derived for linear models. The correction is of particular use when the sample size is small or when the number of fitted parameters is of a moderate to large fraction of the sample size. For linear regression models, the corrected method, called KICc, is an exactly unbiased estimator of a variant of the Kullback symmetric divergence between the true unknown model and the candidate fitted model. Furthermore, KICc is found to provide better model order choice than any other asymptotically efficient methods when applied to autoregressive time series models.
instrumentation and measurement technology conference | 2004
O. Merckel; J.Ch. Bolomey; Gilles Fleury
Specific absorption rate (SAR) designates the electromagnetic power density deposited per unit mass of biological tissues. This paper presents a new approach where SAR calculation for mobile phones is based on a parametric reconstruction of the E-field distribution in the phantom to assess rapid SAR measurements, by means of an ellipsoidal model. The estimation of its parameters is achieved by two ways: using a reduced number of real data points exclusively, and using a combination of real data points E-field extrapolation with a given expansion model.
Journal of The Franklin Institute-engineering and Applied Mathematics | 2004
Gilles Fleury; Alfred O. Hero; Sepidarseh Zareparsi; Anand Swaroop
Most methods for finding interesting gene expression profiles from gene microarray data are based on a single discriminant, e.g. the classical paired t-test. Here a different approach is introduced based on gene ranking according to Pareto depth in multiple discriminants. The novelty of our approach, which is an extension of our previous work on Pareto front analysis (PFA), is that a genes relative rank is determined according to the ordinal theory of multiple objective optimization. Furthermore, the distribution of each genes rank, called Pareto depth, is determined by resampling over the microarray replicates. This distribution is called the Pareto depth sampling distribution (PDSD) and it is used to assess the stability of each ranking. We illustrate and compare the PDSD approach with both simulated and real gene microarray experiments.