Outils pour utilisateurs

Outils du site


opposite_problem

Opposite problem



10H30

Coefficient 1

Presentation of the difficulties arising in the inversion of a physical model. Presentation of the main frameworks and methods addressing the resolution of inverse problems.

Program :


  • Context: case studies; difficulties arising in the inversion of an ill-posed problem (interpretation considering singular values and interpretation in the Fourier domain); information loss, necessity to regularize the problem, concept of prior information.
  • Deterministic approaches: truncated singular value decomposition; estimation by minimization of a cost function.
  • Overview of Bayesian statistics (Bayes’ rule, choice of an estimator, Monte Carlo methods).
  • Stochastic approaches: maximum likelihood estimation – maximum a posteriori estimation – Markovian models – estimation of hyperparameters.


Applications


Laboratory session 1: deterministic methods – application to spike trains deconvolution and tomography (matlab).
Laboratory session 2: stochastic methods – application seismic exploration (matlab).


Back to TU 1 Imaging and Sight Technique in medecine

Back to Innovative Medical Diagnosis and Treatment

opposite_problem.txt · Dernière modification: 2020/10/21 11:27 (modification externe)