Search Talks

In its best understood version, the Steinhardt-Turok cyclic universe contains two crucial ingredients: an unstable field trajectory during the ekpyrotic phase, and the subsequent brane collision corresponding to the crunch/bang transition. These two features act as strong selection principles and determine the broad physical properties of the universe emerging from the bang. As such, they significantly alleviate (and perhaps resolve) the measure problem that is inherent to all cosmological models that produce universes with a range of physical properties.

Galaxy clusters are the biggest gravitationally bound structures in the Universe. Simple features of these objects can help us reconstruct the initial conditions at the Big-Bang and test the fundamental laws of physics.

If the universe is a quantum mechanical system it has a quantum state. This state supplies a probabilistic measure for alternative histories of the universe. During eternal inflation these histories typically develop large inhomogeneities that lead to a mosaic structure on superhorizon scales consisting of homogeneous patches separated by inflating regions. As observers we do not see this structure directly. Rather our observations are confined to a small, nearly homogeneous region within our past light cone. This talk will describe how the probabilities for these observations can be calculated from the probabilities supplied by the quantum state without introducing a further ad hoc measure.

In this talk, I will describe how the collision of Minkowski or crunching bubbles can re-start inflation in a portion of the bubble interior. Consistent with various singularity theorems, such collisions can only seed a lasting inflationary phase with energy density lower than that of the parent vacuum.

The eternal inflation scenario predicts that our observable universe resides inside a single bubble embedded in a vast inflating multiverse. Collisions between bubble universes imprinted in the CMB sky provide a powerful observational test of this idea. I will describe a robust algorithm for non-Gaussian source detection in massive datasets, and present its application to the search for bubble collision signatures in CMB data from WMAP.