when:
| |
(check for the directions here) room Valentin |
who:
Réza Ansari Université Paris-Sud, Laboratoire d’Accélérateur Linéare. More info. | David Mary Université de Nice Sophia Antipolis, Laboratoire J.L. Lagrange | Ben Wandelt Université Pierre et Marie Curie, Institut d’Astrophysique de Paris More info. |
what:
- 14:00-14:35: Réza Ansari (Université Paris-Sud, Laboratoire d’Accélérateur Linéare).
Title: Cosmology : from fundamental questions to computing challenges.
Abstract: Some the great questions in physics and cosmology, such as the nature of dark matter and dark energy will be presented in the first part, followed by a brief overview of the major projects in astrophysics and cosmology. The LSST (Large Synoptic Survey Telescope) and Square Kilometer Array (SKA) radio telescope will be presented. Few of the computing and data management challenges faced by these projects will be discussed in the third and last part of the seminar.
and
- 14:40-15:15: David Mary (Université de Nice Sophia Antipolis, Laboratoire J.L. Lagrange). More info.
Title: Sparse models for image restoration in the perspective of the SKA.
Abstract: This talk will present image restoration methods based on sparse models in the perspective of the new generation of radiotelescopes such as LOFAR. These “software telescopes” are pathfinders for the Square Kilometer Array telescope (SKA), which should be operational in the 2020s. I will in particular present a method
dedicated to the case of faint and diffuse sources buried in the PSF sidelobes of adjacent, bright compact sources.
15:15-15:45 Coffee and a discussion.
- 15:45-16:20: Ben Wandelt (Université Pierre et Marie Curie, Institut d’Astrophysique de Paris). More info.
Title: Probabilistic image reconstruction and power spectrum inference for radio interferometers.
Abstract: I will describe a simple Bayesian approach to simultaneous image reconstruction and covariance inference from radio interferometric data. In simultation comparisons with standard regularization approaches this approach does better in reconstructing the original images. Crucially, it returns uncertainty information about the reconstructions conditioned on the data which allows assessing the reconstruction quality for each pixel and each individual data set.