Young Researchers Conference 2008

I will discuss the various sources of non-Gaussianity (NG) in a class of multi-field models of inflation. I will show that there is both an intrinsic and a local contribution to the NG although they both have the same shape. It is also possible in this class of models that the dominant part to the 3-pt function comes from loop diagrams. These models are of the hybrid type and while they occur naturally in string theory, the conditions for the NG to be important are not generic.

We study various aspects of power suppressed as well as exponentially suppressed corrections in the asymptotic expansion of the degeneracy of quarter BPS dyons in N=4 supersymmetric string theories. In particular we explicitly calculate the power suppressed corrections up to second order and the first exponentially suppressed corrections. We also propose a macroscopic origin of the exponentially suppressed corrections using the quantum entropy function formalism. This suggests a universal pattern of exponentially suppressed corrections to all four dimensional extremal black hole entropies in string theory.

We give an overview of what we have called the \'LQG spinfoam models,\' that provide a spinfoam dynamics for LQG, for arbitrary values of the Barbero-Immirzi parameter in both Lorentzian and Euclidean signatures. The key motivation behind these models was to modify the Barrett-Crane model, by handling more carefully certain constraints, called simplicity constraints, which become second class in the quantum theory. As a result, the kinematics of the models exactly match those of LQG. The goal of this talk is to provide a broad picture of the significance of these models and their basic ideas.

New spin foam models for gravity have been recently proposed to deal with the shortcomings of the Barrett-Crane model. In particular, they draw a closer connection between the Loop Quantum Gravity and the Spin Foam approaches to non perturbative quantum gravity. In this talk, I will present the construction for the case of Lorentzian signature and finite Immirzi parameter. An area operator can be defined and its spectrum agrees with the one defined in LQG. Finally, the amplitude is shown to be finite after a suitable regularization.

More than 40 years ago, Bell ruled out completely local hidden variable models as an explanation for quantum correlations. However, a new type of hidden variable model has recently been brought to light by the work of Leggett. Such a model has both local and non-local parts. Roughly speaking, having a local part means that the measurement outcomes can be guessed with better than 50% success. In this talk, I will explain that there exist quantum correlations for which any hidden variable model must have a trivial local part. I will then discuss how an extension of the original theorem implies that these correlations can be used to enhance the quality of a private random string.

Quantum operations are known to be the most general state transformations that can be applied to parts of compound systems compatibly with the probabilistic structure of quantum mechanics. What about the most general transformations of quantum operations? It turns out that any such general transformation can be realized by a quantum network with an open slot in which the input operation can be inserted, thus programming the resulting circuit. Moreover, one can recursively iterate this construction, generating an infinite hierarchy of admissible transformations and proving their realization within the circuit model of quantum mechanics. These results provide the basis of a new method to optimize quantum networks for information processing tasks, including e.g. gate estimation, discrimination, programming, and cloning. As examples of application, I will present here the optimal quantum networks for estimation of group transformations, for the alignment of reference frames with multiple communication rounds, and for universal cloning of unitary transformations.