Sébastien Gadat

Learn More
We introduce a new model addressing feature selection from a large dictionary of variables that can be computed from a signal or an image. Features are extracted according to an efficiency criterion, on the basis of specified classification or recognition tasks. This is done by estimating a probability distribution P on the complete dictionary, which(More)
Originating in Grenander's pattern theory, the problem of defining appropriate distances between shapes or images and the use of transformation groups to model the variability of natural images is now an active field of research. However, most of the existing results are stated in a deterministic setting while results in a random framework that are(More)
We investigate an important issue of a meta-algorithm for selecting variables in the framework of microarray data. This wrapper method starts from any classification algorithm and weights each variable (i.e. gene) relative to its efficiency for classification. An optimization procedure is then inferred which exhibits important genes for the studied(More)
Microarray technology allows for the monitoring of thousands of gene expressions in various biological conditions, but most of these genes are irrelevant for classifying these conditions. Feature selection is consequently needed to help reduce the dimension of the variable space. Starting from the application of the stochastic meta algorithm " Optimal(More)
The dynamics of the interaction between Cytotoxic T Lymphocytes (CTL) and tumor cells has been addressed in depth, in particular using numerical simulations. However, stochastic mathematical models that take into account the competitive interaction between CTL and tumors undergoing immunoediting, a process of tumor cell escape from immunesurveillance, are(More)
This work investigates the problem of construction of designs for estimation and discrimination between competing linear models. In our framework, the unknown signal is observed with the addition of a noise and only a few evaluations of the noisy signal are available. The model selection is performed in a multi-resolution setting. In this setting, the(More)
This paper considers the problem of estimating a mean pattern in the setting of Grenan-der's pattern theory. Shape variability in a data set of curves or images is modeled by the random action of elements in a compact Lie group on an infinite dimensional space. In the case of observations contaminated by an additive Gaussian white noise, it is shown that(More)
This paper deals with the analysis and the visualization of large graphs. Graphs are convenient widespread data structures that are encountered in a growing number of concrete problems: web, information retrieval, social networks, biological interaction networks… The sizes of these graphs become increasingly large as data acquisition and storage are(More)
In this paper we introduce a new class of diffeomorphic smoothers based on general spline smoothing techniques and on the use of some tools that have been recently developed in the context of image warping to compute smooth diffeomor-phisms. This diffeomorphic spline is defined as the solution of an ordinary differential equation governed by an appropriate(More)
gène) et les données génotypiques (i.e. variables discrètes). Dans le réseau, une interaction entre 2 gènes , c'est à dire le fait que la protéine issue d'un gène active ou inhibe l'expression de l'autre gène, sera représentée par une arête entre ces gènes. Modélisation Nous proposons de modéliser le réseau de la façon suivante : des génotypes des individus(More)