Thu·Oct 5, 2023Thursday·October 5, 2023

1. 01*

   Anisotropy of magnetized quark matter

   Kangkan Goswami🇮🇳 · Dushmanta Sahu🇮🇳 · Jayanta Dey🇮🇳 · Raghunath Sahoo🇮🇳 · Reinhard Stock🇮🇳

   Strong transient magnetic fields are generated in non-central relativistic heavy-ion collisions. These fields induce anisotropy within the strongly interacting medium that, in principle, can affect the thermodynamic properties of the medium. We use the Polyakov loop extended Nambu Jona-Lasinio model to study the quark matter subjected to an external magnetic field at vanishing baryon chemical potential (${\mu }_B$). We have estimated the degree of anisotropy in the speed of sound and isothermal compressibility within the magnetized quark matter as a function of temperature ($T$) and magnetic field ($eB$). This study helps us to understand the extent of directionality generated in the initial stages of non-central collisions while giving us useful information about the system.

   ↳ hep-phhep-exnucl-exnucl-thPRD(2024)·18 citations
2. 02*

   ELUQuant: Event-Level Uncertainty Quantification in Deep Inelastic Scattering

   Cristiano Fanelli🇺🇸 · James Giroux🇺🇸

   We introduce a physics-informed Bayesian Neural Network (BNN) with flow approximated posteriors using multiplicative normalizing flows (MNF) for detailed uncertainty quantification (UQ) at the physics event-level. Our method is capable of identifying both heteroskedastic aleatoric and epistemic uncertainties, providing granular physical insights. Applied to Deep Inelastic Scattering (DIS) events, our model effectively extracts the kinematic variables $x$, $Q^2$, and $y$, matching the performance of recent deep learning regression techniques but with the critical enhancement of event-level UQ. This detailed description of the underlying uncertainty proves invaluable for decision-making, especially in tasks like event filtering. It also allows for the reduction of true inaccuracies without directly accessing the ground truth. A thorough DIS simulation using the H1 detector at HERA indicates possible applications for the future EIC. Additionally, this paves the way for related tasks such as data quality monitoring and anomaly detection. Remarkably, our approach effectively processes large samples at high rates.

   ↳ cs.LGhep-exnucl-exphysics.data-an+1Mach.Learn.Sci.Tech.(2024)·6 citations

* Reconstructed cohort: no mailing for this day survives in the archive. Papers are grouped by their submission times and arXiv's announcement cut-off, assuming announcement without delay; positions follow identifier order. Validated at ~91% exact-day agreement against the archived era.