Birth of the Universe from the Multiverse

It is fair to say that the deepest mystery in our understanding of nature is the birth of our universe. Much of the dilemma over the last decades comes from the extraordinarily small probability that the universe started with the high energy Big Bang as compared to the chance of nucleating any other event. How can Big Bang cosmology be $10^{10^{123}}$ times less likely than nucleating the present cold universe, while accumulating such exquisite agreement with astrophysical data? Why don't we see the other nucleations that, if left to chance, seem to overwhelmingly outnumber us? Here I discuss the point of view that the selection of the initial conditions can be meaningfully addressed only within the framework of the multiverse and that the reason why Big Bang inflation was preferred over other events lies in the quantum dynamics of the landscape of the initial patches. The out-of-equilibrium dynamics selected the 'survivor' universes be born at high energies and the 'terminal' universes at low energies. I briefly review the testable predictions of this theory, in particular the giant void observed in 2007. The second part focuses on the extended framework, in particular a set of postulates needed for defining the multiverse.