Abstract: Cosmic rays with energies up to a few PeV are known to be accelerated within the Milky Way(1,2). Traditionally, it has been presumed that supernova remnants were the main source of these very-high-energy cosmic rays(3,4), but theoretically it is difficult to accelerate protons to PeV energies(5,6) and observationally there simply is no evidence of the remnants being sources of hadrons with energies above a few tens of TeV7,8. One possible source of protons with those energies is the Galactic Centre region(9). Here, we report observations of 1-100 TeV gamma rays coming from the 'Cygnus Cocoon'(10), which is a superbubble that surrounds a region of massive star formation. These gamma rays are likely produced by 10-1,000 TeV freshly accelerated cosmic rays that originate from the enclosed star-forming region Cyg OB2. Until now it was not known that such regions could accelerate particles to these energies. The measured flux likely originates from hadronic interactions. The spectral shape and the emission profile of the Cocoon changes from GeV to TeV energies, which reveals the transport of cosmic particles and historical activity in the superbubble.

Abstract: Simulating nature and in particular processes in particle physics require expensive computations and sometimes would take much longer than scientists can afford. Here, we explore ways to a solution for this problem by investigating recent advances in generative modeling and present a study for the generation of events from a physical process with deep generative models. The simulation of physical processes requires not only the production of physical events, but to also ensure that these events occur with the correct frequencies. We investigate the feasibility of learning the event generation and the frequency of occurrence with several generative machine learning models to produce events like Monte Carlo generators. We study three processes: a simple two-body decay, the processes e(+)e(-)-> Z -> l(+)l(-) and pp -> tt<mml:mo><overbar></mml:mover> including the decay of the top quarks and a simulation of the detector response. By buffering density information of encoded Monte Carlo events given the encoder of a Variational Autoencoder we are able to construct a prior for the sampling of new events from the decoder that yields distributions that are in very good agreement with real Monte Carlo events and are generated several orders of magnitude faster. Applications of this work include generic density estimation and sampling, targeted event generation via a principal component analysis of encoded ground truth data, anomaly detection and more efficient importance sampling, e.g., for the phase space integration of matrix elements in quantum field theories. Here, the authors report buffered-density variational autoencoders for the generation of physical events. This method is computationally less expensive over other traditional methods and beyond accelerating the data generation process, it can help to steer the generation and to detect anomalies.

Abstract: The standard model of particle physics encapsulates our best current understanding of physics at the smallest scales. A fundamental axiom of this theory is the universality of the couplings of the different generations of leptons to the electroweak gauge bosons. The measurement of the ratio of the decay rate of W bosons to t leptons and muons, R(tau/mu), constitutes an important test of this axiom. Using 139 fb(-1) of proton-proton collisions recorded with the ATLAS detector at a centre-of-mass energy of 13 TeV, we report a measurement of this quantity from di-leptonic tt events where the top quarks decay into a W boson and a bottom quark. We can distinguish muons originating from W bosons and those originating from an intermediate t lepton through the muon transverse impact parameter and differences in the muon transverse momentum spectra. The measured value of R(tau/mu) is 0.992 +/- 0.013 [+/- 0.007(stat) +/- 0.011(syst)] and is in agreement with the hypothesis of universal lepton couplings as postulated in the standard model. This is the only such measurement from the Large Hadron Collider, so far, and obtains twice the precision of previous measurements.