Since its formulation by Sir Isaac Newton, the problem of solving the equations of motion for three bodies under their own gravitational force has remained practically unsolved. Currently, the solution for a given initialization can only be found by performing laborious iterative calculations that have unpredictable and potentially infinite computational cost, due to the system's chaotic nature. We show that an ensemble of solutions obtained using an arbitrarily precise numerical integrator can be used to train a deep artificial neural network (ANN) that, over a bounded time interval, provides accurate solutions at fixed computational cost and up to 100 million times faster than a state-of-the-art solver. Our results provide evidence that, for computationally challenging regions of phase-space, a trained ANN can replace existing numerical solvers, enabling fast and scalable simulations of many-body systems to shed light on outstanding phenomena such as the formation of black-hole binary systems or the origin of the core collapse in dense star clusters.

We present the results of a search of the XMM-Newton public data archive for stellar X-ray flares. We find eight flaring sources for which we identify 7 optical counterparts. Three of these sources have distance estimates which allow us to determine their luminosities. Based on the decay time of the flares and their luminosity we derive loop half-lengths of ~2-7x10^10 cm and emission measures of ~10^54 cm^-3: these are similar to values derived for other stellar flaring sources. One of the stars shows two flares in close succession. We discuss the likelihood of this double event being either sympathetic or homologous in nature. A comparison to a pair of similar flares on the Sun suggests that homology is the more likely process driving the flare event.

We report the analysis of the high-resolution soft X-ray spectrum of the nearby F-type star Procyon in the wavelength range from 5 to 175 Angstrom obtained with the Low Energy Transmission Grating Spectrometer (LETGS) on board Chandra and with the Reflection Grating Spectrometers (RGS) and the EPIC-MOS CCD spectrometers on board XMM-Newton. Line fluxes have been measured separately for the RGS and LETGS. Spectra have been fitted globally to obtain self-consistent temperatures, emission measures, and abundances. The total volume emission measure is ~4.1 x 10e50/cm3 with a peak between 1 and 3 MK. No indications for a dominant hot component (T > 4 MK) were found. We present additional evidence for the lack of a solar-type FIP-effect, confirming earlier EUVE results.

XMM-Newton has been performing comprehensive studies of X-ray luminous RS CVn binary systems in its calibration and guaranteed time programs. We present results from ongoing investigations in the context of a systematic study of coronal emission from RS CVns. We concentrate here on coronal abundances and investigate the abundance pattern in RS CVn binaries as a function of activity and average temperature. We find a transition from an Inverse First Ionization Potential (FIP) effect towards an absence of a clear trend (no FIP) in intermediately active RS CVn systems. This scheme fits well into the long-term evolution from an IFIP to a FIP effect found in solar analogs. We further study variations in the elemental abundances during a large flare.

XMM-Newton has been performing comprehensive studies of X-ray bright RS CVn binaries in its Calibration and Guaranteed Time programs. We present results from ongoing investigations in the context of a systematic study of coronal emission from RS CVns. We concentrate in this paper on coronal abundances and investigate the abundance pattern in RS CVn binaries as a function of activity and average temperature. A transition from an Inverse First Ionization Potential (FIP) effect towards an absence of a clear trend is found in intermediately active RS CVn systems. This scheme corresponds well into the long-term evolution from an IFIP to a FIP effect found in solar analogs. We further study variations in the elemental abundances during a large flare.