Search

Results of high-performance optimization of BmnRoot software modules are presented. The BmnRoot package used in the BM@N experiment of the NICA project plays a crucial role in the simulation and event reconstruction so its performance should be maximized to make the data processing efficient. Results of performance analysis on representative testcases are given and bottlenecks are localized. Most suitable approaches to BmnRoot optimization are chosen and numerical estimates of the scalability of the parallelized modules for event reconstruction are presented. [ABSTRACT FROM AUTHOR]

Three deep tracking methods are presented for the BM@N experiment GEM track detector, which differ in their concepts. The first is a two-stage method with data preprocessing by a directional search in the k-d tree to find all possible candidates for tracks, and then use a deep recurrent neural network to classify them by true and ghost tracks. The second end-to-end method used a deep recurrent neural network to extrapolate the initial tracks, similar to the Kalman filter, which learns necessary parameters from the data. The third method implements our new attempt to adapt the neural graph network approach developed in the HEP.TrkX project at CERN to GEM-specific data. The results of applying these three methods to simulated events are presented. [ABSTRACT FROM AUTHOR]

The BM@N experiment (Dubna, JINR) within the NICA complex is a running fixed target experiment. This paper mainly discusses the software support of the experimen and presents the status of the development and testing algorithms to be used for the reconstruction of heavy-ion collisions. Algorithms of reconstruction are described and tested with Monte Carlo input demonstrating a reasonable level of quality assurance. First results on particle identification being a good probe to estimate quality of reconstruction are also mentioned. [ABSTRACT FROM AUTHOR]

We study the performance of an automated hybrid Monte Carlo (HMC) approach for conditional simulation of a recently proposed, single-parameter Gibbs Markov random field. This is based on a modified version of the planar rotator (MPR) model and is used for efficient gap filling in gridded data. HMC combines the deterministic over-relaxation method and the stochastic Metropolis update with dynamically adjusted restriction and performs automatic detection of the crossover to the targeted equilibrium state. We focus on the ability of the algorithm to efficiently drive the system to equilibrium at very low temperatures even with sparse conditioning data. These conditions are the most challenging computationally, requiring extremely long relaxation times if simulated by means of the standard Metropolis algorithm. We demonstrate that HMC has considerable benefits in terms of both computational efficiency and prediction performance of the MPR method. [ABSTRACT FROM AUTHOR]

Evaluation of global thermodynamic properties such as the entropy or the free energy of complex systems featuring a high degree of frustration or disorder is often desirable. Nevertheless, they cannot be measured directly in standard Monte Carlo simulation. Therefore, they are either evaluated indirectly from the directly measured quantities, for example by the thermodynamic integration method (TIM), or by applying more sophisticated simulation methods, such as the Wang-Landau (WL) algorithm, which can directly sample density of states. In the present investigation we compare the performance of the WL and TIM methods for the calculation of the entropy of an Ising antiferromagnetic system on a Kagome lattice – a typical example of a complex spin system with high geometrical frustration resulting in a non-zero residual entropy the value of which is exactly known. It is found that the easier to implement TIM can yield results of comparable accuracy with that of the more involved WL method. [ABSTRACT FROM AUTHOR]