International Journal of Modern Physics A

The Belle II experiment is designed to collect 50 times more data than its predecessor. For a smooth collection of high-quality data, a robust and automated data transport and processing pipeline has been established. We describe the basic software components employed by the high level trigger. It performs a reconstruction of all events using the same algorithms as offline, classifies the events according to physics criteria, and provides monitoring information. The improved system described in this paper has been deployed successfully since 2019.

Classification of processes in heavy-ion collisions in the CBM experiment (FAIR/GSI, Darmstadt) using neural networks is investigated. Fully-connected neural networks and a deep convolutional neural network are built to identify quark–gluon plasma simulated within the Parton-Hadron-String Dynamics (PHSD) microscopic off-shell transport approach for central Au+Au collision at a fixed energy. The convolutional neural network outperforms fully-connected networks and reaches 93% accuracy on the validation set, while the remaining only 7% of collisions are incorrectly classified.

In high energy particle colliders, detectors record millions of points of data during collision events. Therefore, good data analysis depends on distinguishing collisions which produce particles of interest (signal) from those producing other particles (background). Machine learning algorithms in the current times have become popular and useful as the method of choice for such large scale data analysis. In this work, we propose and implement an artificial neural network architecture to achieve the task of identifying precisely the parent particles from all the candidates arising out of track reconstruction from collision data in the future Compressed Baryonic Matter (CBM) experiment. Our framework performs comparably to the existing computational algorithm for this task even with a simple network architecture.

Over the last decade, machine learning algorithms have become standard analysis tools in astroparticle physics, used by a variety of instruments and for an even larger variety of analyses. While a few characteristic patterns can be observed, the portability of established machine learning-based analysis chains from one experiment to another, remains challenging, as instrument-specific prerequisites and adjustments need to be addressed prior to the application. The use Boosted Decision Trees and other tree-based ensemble methods, has been established, but also recently been challenged by the overall success of Deep Neural Networks. Machine learning has been applied for particle selection and parameter reconstruction, as well as for the extraction of energy spectra. This paper aims at summarizing some of the most common approaches on the application of machine learning in astroparticle physics and at providing brief overview on how they have been applied in practice.

Neutrino experiments study the least understood of the Standard Model particles by observing their direct interactions with matter or searching for ultra-rare signals. The study of neutrinos typically requires overcoming large backgrounds, elusive signals, and small statistics. The introduction of state-of-the-art machine learning tools to solve analysis tasks has made major impacts to these challenges in neutrino experiments across the board. Machine learning algorithms have become an integral tool of neutrino physics, and their development is of great importance to the capabilities of next generation experiments. An understanding of the roadblocks, both human and computational, and the challenges that still exist in the application of these techniques is critical to their proper and beneficial utilization for physics applications. This review presents the current status of machine learning applications for neutrino physics in terms of the challenges and opportunities that are at the intersection between these two fields.

On the way towards quantum gravity and the unification of interaction, several ideas have been rejected and avenues avoided because they were perceived as physically unviable. But in the literature there are works in which it was found the contrary, namely that those rejected topics make sense after all. Such topics, reviewed in this paper, are negative energies occurring in higher derivative theories and ultrahyperbolic spaces, ordering ambiguity of operators in curved spaces, the vast landscape of possible compactifications of extra dimensions in string theory, and quantization of a 3-brane in braneworld scenarios.

Lattice measurements provide adequate information to fix the parameters of long-distance effective field theories in Euclidean time. Using such a theory, we examine the analytic continuation of long-distance correlation functions of composite operators at finite temperature from Euclidean to Minkowski space–time. We show through an explicit computation that the analytic continuation of the pion correlation function is possible and gives rise to nontrivial effects. Among them is the possibility, supported by lattice computations of Euclidean correlators, that long distance excitations can be understood in terms of (very massive) pions even at temperatures higher than the QCD crossover temperature.

The experiments at CERN’s Large Hadron Collider use the Worldwide LHC Computing Grid, the WLCG, for its distributed computing infrastructure. Through the distributed workload and data management systems, they provide seamless access to hundreds of grid, HPC and cloud based computing and storage resources that are distributed worldwide to thousands of physicists. LHC experiments annually process more than an exabyte of data using an average of 500,000 distributed CPU cores, to enable hundreds of new scientific results from the collider. However, the resources available to the experiments have been insufficient to meet data processing, simulation and analysis needs over the past five years as the volume of data from the LHC has grown. The problem will be even more severe for the next LHC phases. High Luminosity LHC will be a multiexabyte challenge where the envisaged Storage and Compute needs are a factor 10 to 100 above the expected technology evolution. The particle physics community needs to evolve current computing and data organization models in order to introduce changes in the way it uses and manages the infrastructure, focused on optimizations to bring performance and efficiency not forgetting simplification of operations. In this paper we highlight a recent R&D project related to scientific data lake and federated data storage.

We compute the partition functions of ${𝒩}=1$ gauge theories on $S^2\times {ℝ}_{𝜀}^2$ using supersymmetric localization. The path integral reduces to a sum over vortices at the poles of $S^2$ and at the origin of ${ℝ}_{𝜀}^2$. The exact partition functions allow us to test Seiberg duality beyond the supersymmetric index. We propose the ${𝒩}=1$ partition functions on the $\mathrm{\Omega }$-background, and show that the Nekrasov partition functions can be recovered from these building blocks.

We present a simple method for analytic continuation of harmonic sums near negative and positive integer numbers. We provide a precomputed database for the exact expansion of harmonic sums over a small parameter near these integer numbers, along with MATHEMATICA code, which shows the application of the database for actual problems. We also provide the FORM code that was used to obtain the database mentioned above. The applications of the obtained database for the study of evolution equations in the quantum field theory are discussed.

We consider the nonlinear classical field theory which results from adding to the Maxwell’s Lagrangian the contributions from the weak-field Euler–Heisenberg Lagrangian and a nonuniform part which involves derivatives of the electric and magnetic fields. We focus on the electrostatic case where the magnetic field is set to zero, and we derive the modified Gauss law, resulting in a higher-order differential equation. This equation gives the electric field produced by stationary charges in the higher-order nonlinear electrodynamics. Specializing for the case of a point charge, we investigate the solutions of the modified Gauss law and calculate the correction to the Coulomb law.

In the framework of finite-dimensional Fock space models, for a predefined fixed mean number of particles ${\bar{n}}_k$, it is shown that there is a “large” multidimensional subspace $s_{{\bar{n}}_k}$ of initial pure states, in the space $S$ of all pure states, unitarily evolving to a subspace $S_{{\bar{n}}_k}$ of final pure states which yield ${\bar{n}}_k$. As an example, in particular it follows that the blackbody form of the mean number of particles ${\bar{n}}_k$ does not by itself contradict unitarity of black hole evaporation.

A class of effective field theory called delta-theory, which improves ultraviolet divergences in quantum field theory, is considered. We focus on a scalar model with a quartic self-interaction term and construct the delta theory by applying the so-called delta prescription. We quantize the theory using field variables that diagonalize the Lagrangian, which include a standard scalar field and a ghost or negative norm state. As well known, the indefinite metric may lead to the loss of unitary of the $S$-matrix. We study the optical theorem and check the validity of the cutting equations for three processes at one-loop order, and found suppressed violations of unitarity in the delta coupling parameter of the order of ${\xi }^4$.