Events and Seminars at or with MPE

Special USM Colloquium

Cosmology Seminar

MPP Colloquium

AGN Club

2503 1487334683

Geometry and energy of an AGN-driven outflow

[more]

Munich Joint Astronomy Colloquium

2604 1487334747

Galaxy Cluster Studies with the South Pole Telescope

[more]

Bayes-Forum

1138 1487075852

Surrogate minimization in high dimensions

  • Date: Feb 24, 2017
  • Time: 14:00 - 15:00
  • Speaker: Fabrizia Guglielmetti (ESO)
  • Fabrizia Guglielmetti received a Ph.D. in astronomy from the Ludwig-Maximilians University, Munich, in 2011. The PhD thesis, titled: "Background-Source Separation in astronomical images with Bayesian Probability Theory", was developed under the guidance of Dr. Rainer Fischer and Prof. Volker Dose at the MPI for Plasma Physics. She held positions at the National Institute for Astrophysics (Turin), the Space Telescope Science Institute (Baltimore), MPE and MPA. She is currently working at ESO within the ALMA project. Her main interests are data and image analysis, Bayesian probability theory, interferometry, astrometry and cosmology.
  • Location: MPA
  • Room: Large Seminar Room E 0.11
  • Host: MPA

Abstract: Image interpretation is an ill-posed inverse problem, requiring inference theory methods for a robust solution. Bayesian statistics provides the proper principles of inference to solve the ill-posedness in astronomical images, enabling explicit declaration on relevant information entering the physical models. Furthermore, Bayesian methods require the application of models that are moderately to extremely computationally expensive. Often, the Maximum a Posteriori (MAP) solution is used to estimate the most probable signal configuration (and uncertainties) from the posterior pdf of the signal given the observed data. Locating the MAP solution becomes a numerically challenging problem, especially when estimating a complex objective function defined in an high-dimensional design domain. Therefore, there is the need to utilize fast emulators for much of the required computations. We propose to use Kriging surrogates to speed up optimization schemes, like steepest descent. Results are presented with application on astronomical images, showing the proposed method can effectively search the global optimum. [more]

 
loading content