Geometric numerical integration of stochastic differential problems
seminari CDLab
Raffaele D’Ambrosio
Università degli Studi dell’Aquila
COSA
seminario
QUANDO
3/9/2020
DOVE
Microsoft Teams
REFERENTE
Dimitri Breda
dimitri.breda@uniud.it
DESCRIZIONE
The seminar will take place at 15:00 on Teams: for the link please contact dimitri.breda@uniud.it.
Abstract. The talk will highlight some preliminary principles of stochastic geometric numerical integration, as arisen in the literature of the last few years. In particular, the attention is focused on stochastic differential equations satisfying some characteristic invariance laws. The behaviour of stochastic multistep methods in the preservation of mean-square contractivity will be analyzed. The analysis will also be conveyed to the discretization of stochastic Hamiltonian problems and the numerical preservation of the behaviour of the expected Hamiltonian. The theoretical analysis will also be supported by selected numerical tests.
