arXiv subjects

Wan-Bing He

Publications and source records attributed to Wan-Bing He.

8 recordsLinked to original sources

arXiv2026

Understanding Energy Dependent Hadronic Calorimeter Response from a Machine Learning Perspective

To meet the precision requirements of future high-energy physics experiments, improving the energy resolution of hadronic calorimeters remains a critical challenge. This work presents a systematic investigation of hadronic energy reconstruction using machine learning, highlighting the roles of various signal channels, including scintillation light, Cherenkov light, charged particles, and the full three-dimensional topology of hadronic showers in the energy range up to 10 GeV. Throughout this study, detector effects are not taken into account. Under these conditions, the intrinsic resolution of hadronic showers reaches approximately $(10.8\pm0.3)\% / \sqrt{E/GeV}$ when all signal channels and the full 3D shower information are fully utilized. Compared with the traditional signal-summing approach, machine-learning-based reconstruction can significantly improve energy resolution, even under a limited sampling fraction of 10\%, enhancing it from $(57.6\pm3.7)\%/\sqrt{E/GeV}$ to $(34.1\pm2.8)\%/\sqrt{E/GeV}$. These results highlight the critical importance of both multi-channel information and detailed spatial shower features in hadronic energy reconstruction, and demonstrate the substantial potential of combining high-granularity and dual-readout calorimeter designs with machine-learning-based reconstruction techniques for future experiments.

hep-ex↗

arXiv2026

Neural-network excited states of $A=4$ nuclei and hypernuclei

We present the first variational Monte Carlo study of nuclear and hypernuclear excited states within the neural-network quantum states (NQS) framework. We implement both the overlap penalty (OP) and natural excited state (NES) methods to compute low-lying excitation spectra. To address the spin contamination in hypernuclear calculations, we propose a quantum number targeting (QNT) technique for the OP method. Both the OP-QNT and NES methods can reproduce diagonal observables, such as energies and spatial structures, in excellent agreement with rigorous benchmarks. We further provide, to our knowledge, the first \textit{ab initio} calculation of the $M1$ transition strength for $^{4}_{\Lambda}\mathrm{H}$. The calculated transition strength is consistent with the weak-coupling limit, exhibiting a $\sim$1.3\% suppression. This work demonstrates that NQS can be elevated from ground-state solvers to practical tools for nuclear and hypernuclear spectroscopy.

nucl-th↗

arXiv2025

Machine learning the single-$\Lambda$ hypernuclei with neural-network quantum states

Single-$\Lambda$ hypernuclei are the most straightforward extension of atomic nuclei. A thorough description of baryonic system beyond first-generation quark sector is indispensable for the maturation of nuclear $ab$ $initio$ methods. This study pioneers the application of neural-network quantum states to hypernuclei, with trainable parameters determined by variational Monte Carlo approach (VMC-NQS). In order to reduce the numerical uncertainty and treat the nucleons and hyperons in a unified manner, spinor grouping (SG) method is proposed to analytically integrate out isospin degrees of freedom. A novel spin purification scheme is developed to address the severe spin contamination occurring in standard energy minimization due to the weakly bound characteristic of light single-$\Lambda$ hypernuclei. The energy spectrum of $s$-shell hypernuclei is computed with one-thousandth level accuracy and benchmarked against existing stochastic variational results, showing superior performance. By comparing two different sets of Hamiltonian based on pionless effective field theory (pionless EFT), we choose an optimal model and further carry out calculations of selected $p$-shell charge-symmetric hypernuclei with mass number up to 13, exhibiting satisfactory consistency with experimental results. Our findings underscore the potential of VMC-NQS family in approaching exact solution of few-body systems and the accuracy of pionless EFT in modeling hypernuclei. This is crucial for understanding hyperon-nucleon-nucleon and hyperon-hyperon-nucleon interactions, providing a powerful tool for precisely predicting the properties of multi-strangeness hypernuclei.

nucl-th↗

arXiv2024

Fudan Multi-purpose Active TArget Time Projection Chamber (fMeta-TPC) for Photonnuclear Reaction Experiments

Active Target Time Projection Chambers (AT-TPCs) are state-of-the-art tools in the field of low-energy nuclear physics, particularly suitable for experiments using low-intensity radioactive ion beams or gamma rays. The Fudan Multi-purpose Active Target Time Projection Chamber (fMeta-TPC) with 2048 channels has been developed to study $α$-clustering nuclei. {\fcb In this work, the focus is on the study of the photonuclear reaction with the Laser Compton Scattering (LCS) gamma source, especially for the decay of the highly excited $α$-cluster state.} The design of fMeta-TPC is described and a comprehensive evaluation of its offline performance is performed by ultraviolet (UV) laser and $^{241}$Am $α$ source. The result shows that the intrinsic angular resolution of the detector is within 0.30$^{\circ}$ and has an energy resolution of 6.85\% for 3.0 MeV $α$ particles. The gain uniformity of the detector is about 10\% (RMS/Mean), tested by the $^{55}$Fe X-ray source.

physics.ins-det↗

arXiv2023

High energy nuclear physics meets Machine Learning

Though being seemingly disparate and with relatively new intersection, high energy nuclear physics and machine learning have already begun to merge and yield interesting results during the last few years. It's worthy to raise the profile of utilizing this novel mindset from machine learning in high energy nuclear physics, to help more interested readers see the breadth of activities around this intersection. The aim of this mini-review is to introduce to the community the current status and report an overview of applying machine learning for high energy nuclear physics, to present from different aspects and examples how scientific questions involved in high energy nuclear physics can be tackled using machine learning.

hep-ph↗

arXiv2021

Machine-learning-based identification for initial clustering structure in relativistic heavy-ion collisions

$α$-clustering structure is a significant topic in light nuclei. A Bayesian convolutional neural network (BCNN) is applied to classify initial non-clustered and clustered configurations, namely Woods-Saxon distribution and three-$α$ triangular (four-$α$ tetrahedral) structure for $^{12}$C ($^{16}$O), from heavy-ion collision events generated within a multi-phase transport (AMPT) model. Azimuthal angle and transverse momentum distributions of charged pions are taken as inputs to train the classifier. On multiple-event basis, the overall classification accuracy can reach $95\%$ for $^{12}$C/$^{16}$O + $^{197}$Au events at $\sqrt{S_{NN}} =$ 200 GeV. With proper constructions of samples, the predicted deviations on mixed samples with different proportions of both configurations could be within $5\%$. In addition, setting a simple confidence threshold can further improve the predictions on the mixed dataset. Our results indicate promising and extensive possibilities of application of machine-learning-based techniques to real data and some other problems in physics of heavy-ion collisions.

hep-ph↗

arXiv2020

Nuclear liquid-gas phase transition with machine learning

The machine-learning techniques have shown their capability for studying phase transitions in condensed matter physics. Here, we employ the machine-learning techniques to study the nuclear liquid-gas phase transition. We adopt an unsupervised learning and classify the liquid and gas phases of nuclei directly from the final state raw experimental data of heavy-ion reactions. Based on a confusion scheme which combines the supervised and unsupervised learning, we obtain the limiting temperature of the nuclear liquid-gas phase transition. Its value $9.24\pm0.04~\rm MeV$ is consistent with that obtained by the traditional caloric curve method. Our study explores the paradigm of combining the machine-learning techniques with heavy-ion experimental data, and it is also instructive for studying the phase transition of other uncontrollable systems, like QCD matter.

nucl-th↗

arXiv2018

Collective flows of $^{16}$O + $^{16}$O collisions with $α$-clustering configurations

The main purpose of the present paper is to discuss whether or not the collective flows in heavy-ion collision at Fermi energy can be taken as a tool to investigate the cluster configuration in light nuclei. In practice, within an Extended Quantum Molecular Dynamics model, four $α$-clustering (linear chain, kite, square, and tetrahedron) configurations of $^{16}$O are employed in the initialization, $^{16}$O+$^{16}$O around Fermi energy (40 - 60 MeV$/$nucleon) with impact parameter 1 - 3 fm are simulated, the directed and elliptic flows are analyzed. It is found that collective flows are influenced by the different $α$-clustering configurations, and the directed flow of free protons is more sensitive to the initial cluster configuration than the elliptic flow. Nuclear reaction at Fermi energy can be taken a useful way to study cluster configuration in light nuclei.

nucl-th↗