Filtering Methods for Coupled Inverse Problems

SIAM Journal on Applied Dynamical Systems, vol. 22, pp. 1234–1252

Abstract

Abstract. In many applications it is required to determine a parameters set which is suitable for competing models. To this end, we are interested in ensemble methods to solve multiobjective optimization problems. Here, the ensemble Kalman Filter method is applied and adapted in order to solve coupled inverse nonlinear problems using a weighted function approach. An analysis of the mean field limit of the ensemble method yields an explicit update formula for the weights. Numerical examples show the improved performance of the proposed method.

Authors 2

1. Michaël Herty Aachen Institut für Geometrie und Praktische Mathematik

   RWTH Aachen University

   Affiliation as printed

   Institut für Geometrie und Praktische Mathematik, RWTH Aachen University, Templergraben 55, 52062 Aachen, Germany

2. Elisa Iacomini Aachen Institut für Geometrie und Praktische Mathematik

   RWTH Aachen University

   Affiliation as printed

   Institut für Geometrie und Praktische Mathematik, RWTH Aachen University, Templergraben 55, 52062 Aachen, Germany