Effective parameter dimension in the inverse scattering problem

Thời gian: 09:00 đến 11:00 Ngày 25/11/2016

Địa điểm: C2-714

Báo cáo viên: Marcos A. Capistran

Tóm tắt:

We address a prototype inverse scattering problem in the interface of applied mathematics, statistics, and scientific computing. We pose the acoustic inverse scattering problem in the Bayesian inference perspective and simulate from the corresponding posterior distribution using Markov Chain Monte Carlo. The PDE forward map is implemented using high performance computing methods. We implement a standard Bayesian model selection method to estimate an effective number of Fourier coefficients that may be retrieved from noisy data within a standard formulation.