ASPI - 2012 - Annual activity report

ASPI

ASPI - 2012

Project-Team Aspi

Members

Overall Objectives

Scientific Foundations

Interacting Monte Carlo methods and particle approximation of Feynman–Kac distributions
Statistics of HMM
Multilevel splitting for rare event simulation
Nearest neighbor estimates

Application Domains

Localisation, navigation and tracking
Rare event simulation

New Results

On the length of one–dimensional reactive paths
Long time behavior of piecewise–deterministic Markov processes
Quantitative long time behavior of an ergodic variant of the telegraph process
Total variation estimates for the TCP process
Convergence results for approximate Bayesian computation
Soft level splitting for rare event estimation
Decoding fingerprints using the Markov chain Monte Carlo method
Iterative isotone regression
Detection issues in track–before–detect
Estimation of conflict probability
Minimum volume set for a rare event
Laplace and sequential Monte Carlo methods in Bayesian filtering
Wind–wave modelling
Sequential data assimilation: ensemble Kalman filter vs. particle filter

Partnerships and Cooperations

National initiatives

Dissemination

Scientific animation
Teaching
PhD and habilitation theses
Participation in workshops, seminars, lectures, etc.
Visits and invitations

Bibliography

Major publications
Publications of the year
References in notes

Home | Next next

2012 Project-Team Activity Report

ASPI

Applications of interacting particle systems to statistics

Research centre: Rennes - Bretagne-Atlantique

In partnership with: CNRS, Université Rennes 1, Université Haute Bretagne (Rennes 2)

In collaboration with: Institut de recherche mathématique de Rennes (IRMAR)

Field: Applied Mathematics, Computation and Simulation

Theme: Stochastic Methods and Models

Keywords: Monte Carlo Methods, Markovian Model, Rare Events, Particle Filtering, Tracking, Data Assimilation