Frankfurt Institute for Advanced Studies

About: Frankfurt Institute for Advanced Studies is a facility organization based out in Frankfurt am Main, Germany. It is known for research contribution in the topics: Baryon & Quark–gluon plasma. The organization has 798 authors who have published 2733 publications receiving 82799 citations. The organization is also known as: FIAS.

Preference elicitation and inverse reinforcement learning

Abstract: We state the problem of inverse reinforcement learning in terms of preference elicitation, resulting in a principled (Bayesian) statistical formulation. This generalises previous work on Bayesian inverse reinforcement learning and allows us to obtain a posterior distribution on the agent's preferences, policy and optionally, the obtained reward sequence, from observations. We examine the relation of the resulting approach to other statistical methods for inverse reinforcement learning via analysis and experimental results. We show that preferences can be determined accurately, even if the observed agent's policy is sub-optimal with respect to its own preferences. In that case, significantly improved policies with respect to the agent's preferences are obtained, compared to both other methods and to the performance of the demonstrated policy.

Cooling of hypernuclear compact stars

Abstract: We study the thermal evolution of hypernuclear compact stars constructed from covariant density functional theory of hypernuclear matter and parameterizations which produce sequences of stars containing two-solar-mass objects. For the input in the simulations, we solve the Bardeen-Cooper-Schrieffer gap equations in the hyperonic sector and obtain the gaps in the spectra of $\Lambda$, $\Xi^0$ and $\Xi^-$ hyperons. For the models with masses $M/M_{\odot} \ge 1.5$ the neutrino cooling is dominated by hyperonic direct Urca processes in general. In the low-mass stars the $(\Lambda p)$ plus leptons channel is the dominant direct Urca process, whereas for more massive stars the purely hyperonic channels $(\Sigma^-\Lambda)$ and $(\Xi^-\Lambda)$ are dominant. Hyperonic pairing strongly suppresses the processes on $\Xi^-$s and to a lesser degree on $\Lambda$s. We find that intermediate-mass $1.5 \le M/M_{\odot} \le 1.8$ models have surface temperatures which lie within the range inferred from thermally emitting neutron stars, if the hyperonic pairing is taken into account. Most massive models with $M/M_{\odot} \simeq 2$ may cool very fast via the direct Urca process through the $(\Lambda p)$ channel because they develop inner cores where the $S$-wave pairing of $\Lambda$s and proton is absent.

QCD sum rules for magnetically induced mixing between ηc and J/ψ.

Abstract: We investigate the properties of charmonia in strong magnetic fields by using QCD sum rules. We show how to implement the mixing effects between η(c) and J/ψ on the basis of field-theoretical approaches, and then show that the sum rules are saturated by the mixing effects with phenomenologically determined parameters. Consequently, we find that the mixing effects are the dominant contribution to the mass shifts of the static charmonia in strong magnetic fields.

Continuum versus discrete model: a comparison for multicellular tumour spheroids

Abstract: We study multicellular tumour spheroids with a continuum model based on partial differential equations (PDEs). The model includes viable and necrotic cell densities, as well as oxygen and glucose concentrations. Viable cells consume nutrients and become necrotic below critical nutrient concentrations. Proliferation of viable cells is contact-inhibited if the total cellular density locally exceeds volume carrying capacity. The model is discussed under the assumption of spherical symmetry. Unknown model parameters are determined by simultaneously fitting the cell number to several experimental growth curves for different nutrient concentrations. The outcome of the PDE model is compared with an analogous off-lattice agent-based model for tumour growth. It turns out that the numerically more efficient PDE model suffices to explain the macroscopic growth data. As in the agent-based model, we find that the experimental growth curves are only reproduced when a necrotic core develops. However, evaluation of morphometric properties yields differences between the models and the experiment.