Search

In December 2023, the European Parliament provisionally agreed on the EU AI Act. This unprecedented regulatory framework for AI systems lays out guidelines to ensure the safety, legality, and trustworthiness of AI products. This paper presents a methodology for interpreting the EU AI Act requirements for high-risk AI systems by leveraging product quality models. We first propose an extended product quality model for AI systems, incorporating attributes relevant to the Act not covered by current quality models. We map the Act requirements to relevant quality attributes with the goal of refining them into measurable characteristics. We then propose a contract-based approach to derive technical requirements at the stakeholder level. This facilitates the development and assessment of AI systems that not only adhere to established quality standards, but also comply with the regulatory requirements outlined in the Act for high-risk (including safety-critical) AI systems. We demonstrate the applicability of this methodology on an exemplary automotive supply chain use case, where several stakeholders interact to achieve EU AI Act compliance., Comment: To be published in: 2024 IEEE Conference on Artificial Intelligence (CAI 2024)

Cosmic rays (CRs) heavily impact the chemistry and physics of cold and dense star-forming regions. However, characterising their ionisation rate is still challenging from an observational point of view. In the past, a few analytical formulas have been proposed to infer the cosmic-ray ionization rate $\zeta_2$ from molecular line observations. These have been derived from the chemical kinetics of the involved species, but they have not been validated using synthetic data processed with a standard observative pipeline. We aim to bridge this gap. We perform the radiative transfer on a set of three-dimensional magneto-hydrodynamical simulations of prestellar cores, exploring different initial $\zeta_2$, evolutionary stages, types of radiative transfer (e.g. assuming local-thermodynamic-equilibrium conditions), and telescope responses. We then compute the column densities of the involved tracers to determine $\zeta_2$, using, in particular, the equation proposed by Bovino et. al (2020) and by Caselli et al. (1998) both used nowadays. Our results confirm that the method of Bovino et al. (2020) accurately retrieves the actual $\zeta_2$ within a factor of $2-3$, in the physical conditions explored in our tests. Since we also explore a non-local thermodynamic equilibrium radiative transfer, this work indirectly offers insights into the excitation temperatures of common transitions at moderate volume densities ($n\approx 10^5 \, \rm cm^{-3}$). We have also performed a few tests using the formula proposed by Caselli et al. (1998), which overestimates the actual $\zeta_2$ by at least two orders of magnitudes. We also consider a new derivation of this method, which, however, still leads to large overestimates., Comment: Accepted for publication in A&A (15th Feb.)

Mechanical properties are essential for understanding natural and induced deformational behavior of geological formations. Brittleness characterizes energy dissipation rate and strain localization at failure. Brittleness has been investigated in hydrocarbon-bearing mudrocks in order to quantify the impact of hydraulic fracturing on the creation of complex fracture networks and surface area for reservoir drainage. Typical well logging correlations associate brittleness with carbonate content or dynamic elastic properties. However, an index of rock brittleness should involve actual rock failure and have a consistent method to quantify it. Here we present a systematic method to quantify mudrock brittleness based on micro-mechanical measurements from the scratch test. Brittleness is formulated as the ratio of energy associated with brittle failure to the total energy required to perform a scratch. Soda lime glass and polycarbonate are used for comparison to identify failure in brittle and ductile mode and validate the developed method. Scratch testing results on mudrocks indicate that it is possible to use the recorded transverse force to estimate brittleness. Results show that tested samples rank as follows in increasing degree of brittleness: Woodford, Eagle Ford, Marcellus, Mancos, and Vaca Muerta. Eagle Ford samples show mixed ductile/brittle failure characteristics. There appears to be no definite correlation between micro-scratch brittleness and quartz or total carbonate content. Dolomite content shows a stronger correlation with brittleness than any other major mineral group. The scratch brittleness index correlates positively with increasing Young's modulus and decreasing Poisson's ratio, but shows deviations in rocks with distinct porosity and with stress-sensitive brittle/ductile behavior (Eagle Ford).

WR 140 is a long-period, highly eccentric Wolf-Rayet star binary system with exceptionally well-determined orbital and stellar parameters. Bright, variable X-ray emission is generated in shocks produced by the collision of the winds of the WC7pd+O5.5fc component stars. We discuss the variations in the context of the colliding-wind model using broad-band spectrometry from the RXTE, SWIFT, and NICER observatories obtained over 20 years and nearly 1000 observations through 3 consecutive 7.94-year orbits including 3 periastron passages. The X-ray luminosity varies as expected with the inverse of the stellar separation over most of the orbit: departures near periastron are produced when cooling shifts to excess optical emission in CIII $\lambda5696$ in particular. We use X-ray absorption to estimate mass-loss rates for both stars and to constrain the system morphology. The absorption maximum coincides closely with inferior conjunction of the WC star and provides evidence of the ion-reflection mechanism that underlies the formation of collisionless shocks governed by magnetic fields probably generated by the Weibel instability. Comparisons with K-band emission and HeI $\lambda$10830 absorption show that both are correlated after periastron with the asymmetric X-ray absorption. Dust appears within a few days of periastron suggesting formation within shocked gas near the stagnation point. X-ray flares seen in $\eta$ Carinae have not occurred in WR 140, suggesting the absence of large-scale wind inhomogeneities. Relatively constant soft emission revealed during the X-ray minimum is probably not from recombining plasma entrained in outflowing shocked gas., Comment: 52 pages, 17+1 figures, Accepted for publication in The Astrophysical Journal 23 August 2021

The Vista Variables in the V\'ia L\'actea (VVV) ESO Public Survey is an ongoing time-series, near-infrared (IR) survey of the Galactic bulge and an adjacent portion of the inner disk, covering 562 square degrees of the sky, using ESO's VISTA telescope. The survey has provided superb multi-color photometry in 5 broadband filters ($Z$, $Y$, $J$, $H$, and $K_s$), leading to the best map of the inner Milky Way ever obtained, particularly in the near-IR. The main variability part of the survey, which is focused on $K_s$-band observations, is currently underway, with bulge fields having been observed between 31 and 70 times, and disk fields between 17 and 36 times. When the survey is complete, bulge (disk) fields will have been observed up to a total of 100 (60) times, providing unprecedented depth and time coverage. Here we provide a first overview of stellar variability in the VVV data, including examples of the light curves that have been collected thus far, scientific applications, and our efforts towards the automated classification of VVV light curves., Comment: Presented at the "40 Years of Variable Stars: A Celebration of Contributions by Horace A. Smith" conference (arXiv:1310.0149). 49 pages, 28 figures