Partenaires

CNRS
Logo tutelle
Logo tutelle
Logo tutelle


Rechercher

Sur ce site

Sur le Web du CNRS


Accueil du site > Séminaires > Probabilités Statistiques et réseaux de neurones > Solutions of SDEs as saddle points of mini-max problems

Vendredi 10 juin 2005, à 10h

Solutions of SDEs as saddle points of mini-max problems

Istvan Gyongy (Edinburgh University, Grande Bretagne)

Résumé : Adapting an idea of Nicolai Krylov we show that SDEs can be associated with mini-max problems in suitable infinite dimensional spaces. More precisely, if the coefficients of an SDE satisfy the so called monotonicity condition, then one can construct a mini-max problem such that its saddle point is the solution of the given SDE. This connection between SDEs and mini-max problems can be useful in many ways. In a recent joint paper with Annie Millet we make use of the corresponding construction for stochastic PDEs to obtain results on numerical approximations. In the talk we show that the existence of a (strong) solution of the SDE follows directly from a suitable mini-max theorem. This has been observed in recent joint work with Shirley Chen.

Dans la même rubrique :