Multiparton interactions in pp collisions from machine learning-based regression

Abstract

Multiparton interactions (MPI) in pp collisions have attracted the attention of the heavy-ion community since they can help to elucidate the origin of collectivelike effects discovered in small collision systems at the LHC. In this work, we report that in PYTHIA8.244, the charged-particle production in events with a large number of MPI (N mpi) normalized to that obtained in minimum-bias pp collisions shows interesting features. After the normalization to the corresponding hN mpi i, the ratios as a function of p T exhibit a bump at p T ≈ 3 GeV=c; and for higher p T (> 8 GeV=c), the ratios are independent of N mpi. While the size of the bump increases with increasing N mpi , the behavior at high p T is expected from the "binary scaling" (partonparton interactions), which holds given the absence of any parton-energy loss mechanism in PYTHIA. The bump at intermediate p T is reminiscent of the Cronin effect observed for the nuclear modification factor in p-Pb collisions. In order to unveil these effects in data, we propose a strategy to construct an event classifier sensitive to MPI using machine learning-based regression. The study is conducted using TMVA, and the regression is performed with boosted decision trees (BDT). Event properties like forward charged-particle multiplicity, transverse spherocity and the average transverse momentum (hp T i) are used for training. The kinematic cuts are defined in accordance with the ALICE detector capabilities. For the validation of the method and to find possible model dependence, we also compare the results from PYTHIA8.244 with HERWIG7.1. In addition, we also report that if we apply the trained BDT on existing (INEL > 0) pp data, i.e., events with at least one primary charged-particle within jηj < 1, the average number of MPI in pp collisions at ffiffi ffi s p ¼ 5.02 and 13 TeV are 3.76 AE 1.01 and 4.65 AE 1.01, respectively.

Multiparton Interactions in pp collisions from Machine Learning

“…We obtain a regression value which is above unity, therefore, our results support the presence of MPI in pp collisions. We also observe a modest energy dependence, which is similar to that predicted by PYTHIA 8 [7]. A similar finding has been discussed in Ref.…”

“…For instance, color reconnection and MPI can mimic radial flow patterns in pp collisions [6]. In this direction, we have proposed the extraction of the MPI activity from minimum-bias pp data using Machine Learning (ML) methods [7,8]. In this contribution, we summarize the main results including the multiplicity dependence of the average number of MPI extracted from the available ALICE data at the LHC [3,9].…”

Particle production as a function of the underlying event in pp collisions simulated with PYTHIA 8

“…3 exhibit an opposite behaviour: only a weak dependence on the min T event classes is seen. However, the position of its maximum is slightly shifted to higher T values in events with large min T , which is expected in PYTHIA 8 events with large mpi and color reconnection [25][26][27].…”

Estimation of Impact Parameter and Transverse Spherocity in heavy-ion collisions at the LHC energies using Machine Learning

“…Recently, machine learning techniques have led to a range of numerous developments in the field of highenergy physics (HEP) along with in different fields of physics [13][14][15][16][17][18][19]. For several years different machine learning algorithms have been used to determine the impact parameter [20][21][22][23][24].…”