B modes and component separation. The Gibbs sampling solution.
Detecting B-modes due to inflationary gravitational waves in the polarization of the Cosmic Microwave Background is one of the greatest goals of modern observational cosmology. Many experiments have been designed to detect this weak signal, and upper limits are improving year by year. In fact, while B-modes have now been clearly observed, these are not of inflationary origin, but rather induced by late time effects. At large scales, the signal is most probably dominated by the galactic dust, whereas at small scale, it is due to gravitational lensing. In this talk I will give a general introduction to the field, and show some recent observational highlights. I will then review the problem of component separation, and present a recently developed method for joint estimation of cosmological parameters and astrophysical foregrounds. Finally, I will discuss how this may be applied to observations from the SPIDER experiment.