Till innehåll på sidan

Johan Westerborn: On particle-based online smoothing and parameter inference in general state-space models

Tid: On 2017-10-18 kl 15.15

Plats: Room 306, House 6, Kräftriket, Department of Mathematics, Stockholm University

Medverkande: Johan Westerborn (KTH)

Exportera till kalender

Abstract: In this talk I will go over a few topics of my thesis. The main theme of the thesis revolves around the particle-based, rapid, incremental smoother (PaRIS) algorithm for efficient particle-based online smoothing of function of additive form. New algorithms for efficient parameter estimation and online estimation of the marginal smoothing distributions can be derived from the PaRIS algorithm, in this talk I will present some of these algorithms.