- Nourou Sylla (MISTIS) Supervised classification with mixture models: Application to medical diagnostic.
- Laurent Gilquin (STEEP) Stochastic tools for the analysis of the TRANUS model (integrated land-use and transport model).
TRANUS is a LUTI model simulating the location of economic activities, the land market and the transport network on a given territory. The first objective is to develop methodological tools of stochastic nature to calibrate and analyze TRANUS. The second objective is to construct a methodology based on those tools allowing decision-makers to exploit TRANUS properties to propose urban development policies. I will present a first method to calibrate the land-use submodel of TRANUS, method I developped during my internship of the last six months. Then the modifications and further improvements i will work on.
- Alessandro Chiancone (MISTIS) Dimension reduction.
- Christine Bakhous (MISTIS) Parsimonious encoding models for brain activity measured by functional MRI.
Functional magnetic resonance imaging (fMRI) is a noninvasive technique allowing the study of brain activity via the measurement of hemodynamic changes. Recently, a joint detection-estimation (JDE) framework was developed and relies on both (1) the brain activity detection and (2) the hemodynamic response function estimation, two steps that are generally addressed in a separate way. The JDE approach is a parcel-based model that alternates (1) and (2) on each parcel successively. The JDE analysis assumes that all delivered stimuli (e.g. visual, auditory, etc.) possibly generate a response everywhere in the brain although activation is likely to be induced by only some of them in specific brain areas. Inclusion of irrelevant events may degrade the results. Since the relevant conditions or stimulus types can change between different brain areas, a model selection procedure will be computationally expensive. Furthermore, criteria are not always available to select the relevant conditions prior to activation detection, especially in pathological cases. The goal of this work is to develop a JDE extension allowing an automatic selection of the relevant conditions according to the brain activity they elicit. This condition selection is done simultaneously to the analysis and adaptively through the different brain areas. Analysis on simulated and real datasets illustrate the ability of our model to select the relevant conditions and its interest compare to the standard JDE analysis.
- Meryam Krit (FIGAL) Goodness-of-fit tests and model selection for recurrent events in reliability.
- Charlotte Dion (SAM) New strategies for nonparametric estimation in linear mixed models.
- Gildas Mazo (MISTIS) Contributions to copula modeling in high dimension.
The need for modeling multivariate distributions is present in every scientific domain. To meet this need, the concept of copulas has been proved to be a powerful tool even though the construction of copulas in high dimension is challenging. In this communication, we present new high dimensional copulas.
La demi-journée a lieu à 15h30 dans la salle 1 de la tour IRMA, 51 rue des Mathématiques sur le campus. Elle est ouverte à tous.