25/06/2020 @ 10h30 – 11h30 – A modification of quasi-Newton’s methods helping to avoid saddle points Résumé: We propose a new modification of quasi-Newton’s methods, namely New Q-Newton’s method. The main idea is to approximate the Hessian matrix by a invertible (not necessarily positive definite) symmetric matrix, then update the Newton step by the difference of the projections of the standard […]
25/06/2020 @ 10h30 – 11h30 – A modification of quasi-Newton’s methods helping to avoid saddle points Résumé: We propose a new modification of quasi-Newton’s methods, namely New Q-Newton’s method. The main idea is to approximate the Hessian matrix by a invertible (not necessarily positive definite) symmetric matrix, then update the Newton step by the difference of the projections of the standard […]
11/06/2020 @ 10h30 – 11h30 – Safety Management Systems : une réelle voie d’amélioration de la sécurité aéronautique aujourd’hui ? Résumé Depuis le début des années 2000, l’aviation civile adopte progressivement les Safety Management Systems comme voie d’amélioration de la sécurité. Il y a quelques mois encore, suite au rapport d’enquête préliminaire sur les deux accidents du Boeing 737 MAX, la […]
27/02/2020 @ 10h30 – 11h30 – Evolution of daily temperature due to global change Abstract We consider the problem of estimating non-stationary temperature normals (the mean), at a daily timescale, which take into account deformations due to the global change of the annual cycle. In a second step, we consider not only the mean, but the whole distribution. Thus, we will […]
27/02/2020 @ 10h30 – 11h30 – Evolution of daily temperature due to global change Abstract We consider the problem of estimating non-stationary temperature normals (the mean), at a daily timescale, which take into account deformations due to the global change of the annual cycle. In a second step, we consider not only the mean, but the whole distribution. Thus, we will […]
23/01/2020 @ 10h30 – 11h30 – Multivariate Student versus Multivariate Gaussian Regression Models with Applications to Finance and Political Economy Abstract For modeling multivariate possibly heavy-tailed data, we compare the multivariate normal regression model (N) with two versions of the multivariate Student regression model: the independent multivariate Student (IT) and the uncorrelated multivariate Student (UT). We recall some facts about these […]
23/01/2020 @ 10h30 – 11h30 – Multivariate Student versus Multivariate Gaussian Regression Models with Applications to Finance and Political Economy Abstract For modeling multivariate possibly heavy-tailed data, we compare the multivariate normal regression model (N) with two versions of the multivariate Student regression model: the independent multivariate Student (IT) and the uncorrelated multivariate Student (UT). We recall some facts about these […]
12/12/2019 @ 10h30 – 11h30 – Feuilletages des variétés statistiques Résumé: une variété statistique est de façon formelle une variété riemannienne munie de deux connexions duales. Cet objet abstrait apparaît de façon naturelle lorsque l’on étudie des familles paramétrées de densités et formalise certaines procédures courantes en estimation. Dans cet exposé nous aborderons de la façon la plus intuitive possible des […]
12/12/2019 @ 10h30 – 11h30 – Foliation of statistical manifold Summary: A statistical manifold is formally a Riemannian manifold with two dual connections. This abstract object appears naturally when we study parameterized families of densities and formalize some common estimation procedures. In this paper we will address as intuitively as possible the structural questions relating to statistical manifolds and in particular […]
28/11/2019 @ 10h30 – 11h30 – Big Data and Algorithm Transparency