"The hunt for (almost) perfect fluid"

will be given by Dr. Pasi Huovinen (University of Wroclaw, Poland).

Abstract of the talk:

The fundamental building blocks of matter, quarks and gluons, are always confined to form hadrons. However, we expect that in extremely large temperatures and densities this confinement would be broken and quarks and gluons would move freely forming so called quark matter or quark-gluon plasma. It is believed that such a state of matter did exist a few milliseconds after the big bang, and that it has been recreated in ultra-relativistic heavy-ion collisions of large nuclei in the experiments at Brookhaven National Laboratory and CERN. It looks like that the matter created in these collisions has extraordinary properties like such a low kinematic viscosity that it has been described as perfect fluid. In this talk we will describe how we have come to believe that quark-gluon plasma has been created in the heavy-ion collisions, and our attempts to evaluate its hydrodynamical properties.

The second talk entitled

"Constraining the parameters of a heavy ion collision model using Bayesian statistics"

will be given by Dr. Jussi Auvinen (Duke University, USA).

Abstract of the talk:

Model-to-data comparison in the field of relativistic heavy ion collisions is a complicated task, as the models typically have multiple input parameters, which need to be simultaneously tuned to an even larger set of experimentally measured observables. When performing such a multidimensional parameter fit, it would be important to also be able to quantify the uncertainties related to the determined best-fit values, as there could be various parameter combinations, which describe the experimental data equally well.

Bayesian statistics approach has the advantage of providing both the optimal values and their uncertainties simultaneously, in the form of a posteriori probability distribution of the input parameter space. The peak of the posterior distribution indicates the most probable input parameter combination, given the experimental data, while the width of the distribution reflects the associated uncertainty.

In this talk we present an example case of using Bayesian statistics for finding the optimal values for the input parameters of a hadron transport + relativistic viscous hydrodynamics hybrid model for Au+Au collisions at center-of-mass energies 19.6, 39, and 62.4 GeV. While some parameters are well constrained by the present data set, others still have large uncertainties regarding the optimal value. In addition, the optimal parameter values are found to depend on collision energy.