arXiv subjects

Jie Meng

Publications and source records attributed to Jie Meng.

At least 73 records · Page 4Linked to original sources

arXiv2021

Dibaryon with highest charm number near unitarity from lattice QCD

A pair of triply charmed baryons, $Ω_{ccc}Ω_{ccc}$, is studied as an ideal dibaryon system by (2+1)-flavor lattice QCD with nearly physical light-quark masses and the relativistic heavy quark action with the physical charm quark mass. The spatial baryon-baryon correlation is related to their scattering parameters on the basis of the HAL QCD method. The $Ω_{ccc}Ω_{ccc}$ in the ${^1S_0}$ channel taking into account the Coulomb repulsion with the charge form factor of $Ω_{ccc}$ leads to the scattering length $a^{\rm C}_0\simeq -19~\text{fm}$ and the effective range $r^{\rm C}_{\mathrm{eff}}\simeq 0.45~\text{fm}$. The ratio $r^{\rm C}_{\mathrm{eff}}/a^{\rm C}_0 \simeq -0.024$, whose magnitude is considerably smaller than that of the dineutron ($-0.149$), indicates that $Ω_{ccc}Ω_{ccc}$ is located in the unitary regime.

hep-lat↗

arXiv2021

Relativistic chiral description of the $^1S_0$ nucleon-nucleon scattering

Recently, a relativistic chiral nucleon-nucleon interaction is formulated up to leading order which provides a good description of the phase shifts of $J\leq1$ partial waves [Chin. Phys. C 42 (2018) 014103]. Nevertheless, a separable regulator function that is not manifestily covariant was used in solving the relativistic scattering equation. In the present work, we first propose a covariant and separable form factor to regularize the kernel potential and then apply it to study the simplest but most challenging $^1S_0$ channel which features several low-energy scales. In addition to being self-consistent, we show that the resulting relativistic potential can describe quite well the unique features of the $^1S_0$ channel at leading order, in particular the pole position of the virtual bound state and the zero amplitude at the scattering momentum $\sim 340$ MeV, indicating that the relativistic formulation might be more natural from the point of view of effective field theories.

nucl-th↗

arXiv2021

Coverage Guided Testing for Recurrent Neural Networks

Recurrent neural networks (RNNs) have been applied to a broad range of applications, including natural language processing, drug discovery, and video recognition. Their vulnerability to input perturbation is also known. Aligning with a view from software defect detection, this paper aims to develop a coverage guided testing approach to systematically exploit the internal behaviour of RNNs, with the expectation that such testing can detect defects with high possibility. Technically, the long short term memory network (LSTM), a major class of RNNs, is thoroughly studied. A family of three test metrics are designed to quantify not only the values but also the temporal relations (including both step-wise and bounded-length) exhibited when LSTM processing inputs. A genetic algorithm is applied to efficiently generate test cases. The test metrics and test case generation algorithm are implemented into a tool TestRNN, which is then evaluated on a set of LSTM benchmarks. Experiments confirm that TestRNN has advantages over the state-of-art tool DeepStellar and attack-based defect detection methods, owing to its working with finer temporal semantics and the consideration of the naturalness of input perturbation. Furthermore, TestRNN enables meaningful information to be collected and exhibited for users to understand the testing results, which is an important step towards interpretable neural network testing.

cs.LG↗

arXiv2021

Nuclear matter in relativistic Brueckner-Hartree-Fock theory with Bonn potential in the full Dirac space

Starting from the Bonn potential, relativistic Brueckner-Hartree-Fock (RBHF) equations are solved for nuclear matter in the full Dirac space, which provides a unique way to determine the single-particle potentials and avoids the approximations applied in the RBHF calculations in the Dirac space with positive-energy states (PESs) only. The uncertainties of the RBHF calculations in the Dirac space with PESs only are investigated, and the importance of the RBHF calculations in the full Dirac space is demonstrated. In the RBHF calculations in the full Dirac space, the empirical saturation properties of symmetric nuclear matter are reproduced, and the obtained equation of state agrees with the results based on the relativistic Green's function approach up to the saturation density.

nucl-th↗

arXiv2021

Geometric means of quasi-Toeplitz matrices

We study means of geometric type of quasi-Toeplitz matrices, that are semi-infinite matrices $A=(a_{i,j})_{i,j=1,2,\ldots}$ of the form $A=T(a)+E$, where $E$ represents a compact operator, and $T(a)$ is a semi-infinite Toeplitz matrix associated with the function $a$, with Fourier series $\sum_{\ell=-\infty}^{\infty} a_\ell e^{\mathfrak i \ell t}$, in the sense that $(T(a))_{i,j}=a_{j-i}$. If $a$ is \rv\ and essentially bounded, then these matrices represent bounded self-adjoint operators on $\ell^2$. We consider the case where $a$ is a continuous function, where quasi-Toeplitz matrices coincide with a classical Toeplitz algebra, and the case where $a$ is in the Wiener algebra, that is, has absolutely convergent Fourier series. We prove that if $a_1,\ldots,a_p$ are continuous and positive functions, or are in the Wiener algebra with some further conditions, then means of geometric type, such as the ALM, the NBMP and the Karcher mean of quasi-Toeplitz positive definite matrices associated with $a_1,\ldots,a_p$, are quasi-Toeplitz matrices associated with the geometric mean $(a_1\cdots a_p)^{1/p}$, which differ only by the compact correction. We show by numerical tests that these operator means can be practically approximated.

math.NA↗

arXiv2020

Accelerating High-Order Stencils on GPUs

Stencil computations are widely used in HPC applications. Today, many HPC platforms use GPUs as accelerators. As a result, understanding how to perform stencil computations fast on GPUs is important. While implementation strategies for low-order stencils on GPUs have been well-studied in the literature, not all of proposed enhancements work well for high-order stencils, such as those used for seismic modeling. Furthermore, coping with boundary conditions often requires different computational logic, which complicates efficient exploitation of the thread-level parallelism on GPUs. In this paper, we study high-order stencils and their unique characteristics on GPUs. We manually crafted a collection of implementations of a 25-point seismic modeling stencil in CUDA and related boundary conditions. We evaluate their code shapes, memory hierarchy usage, data-fetching patterns, and other performance attributes. We conducted an empirical evaluation of these stencils using several mature and emerging tools and discuss our quantitative findings. Among our implementations, we achieve twice the performance of a proprietary code developed in C and mapped to GPUs using OpenACC. Additionally, several of our implementations have excellent performance portability.

cs.DC↗

arXiv2020

CasGCN: Predicting future cascade growth based on information diffusion graph

Sudden bursts of information cascades can lead to unexpected consequences such as extreme opinions, changes in fashion trends, and uncontrollable spread of rumors. It has become an important problem on how to effectively predict a cascade' size in the future, especially for large-scale cascades on social media platforms such as Twitter and Weibo. However, existing methods are insufficient in dealing with this challenging prediction problem. Conventional methods heavily rely on either hand crafted features or unrealistic assumptions. End-to-end deep learning models, such as recurrent neural networks, are not suitable to work with graphical inputs directly and cannot handle structural information that is embedded in the cascade graphs. In this paper, we propose a novel deep learning architecture for cascade growth prediction, called CasGCN, which employs the graph convolutional network to extract structural features from a graphical input, followed by the application of the attention mechanism on both the extracted features and the temporal information before conducting cascade size prediction. We conduct experiments on two real-world cascade growth prediction scenarios (i.e., retweet popularity on Sina Weibo and academic paper citations on DBLP), with the experimental results showing that CasGCN enjoys a superior performance over several baseline methods, particularly when the cascades are of large scale.

cs.SI↗

arXiv2020

Deformed relativistic Hartree-Bogoliubov theory in continuum with point coupling functional: examples of even-even Nd isotopes

The aim of this work is to develop the deformed relativistic Hartree-Bogoliubov theory in continuum (DRHBc) theory based on the point-coupling density functionals and extend it to provide a unified description for all even-even nuclei in the nuclear chart by overcoming all possible challenges. The nuclear superfluidity is considered via Bogoliubov transformation. Densities and potentials are expanded in terms of Legendre polynomials to include the axial deformation degrees of freedom. Sophisticated relativistic Hartree-Bogoliubov equations in coordinate space are solved in the DiracWoods-Saxon basis to consider the continuum effects. Numerical checks are performed from light nuclei to heavy nuclei. The techniques to construct the DRHBc mass table for even-even nuclei are explored. The DRHBc theory is extended to study heavier nuclei beyond magnesium isotopes. Taking Nd isotopes as examples, the experimental binding energies, two-neutron separation energies, quadrupole deformations, and charge radii are reproduced rather well. The deformation and continuum play essential roles in the description of nuclear masses and prediction of drip-line nuclei. By examining the single-particle levels in the canonical basis and their contributions to the total density, the thickness of the neutron skin, the particles number in continuum, and the Coulomb barrier, the exotic structures including the neutron skin and the proton radioactivity are predicted.

nucl-th↗

arXiv2020

Minimod: A Finite Difference solver for Seismic Modeling

This article introduces a benchmark application for seismic modeling using finite difference method, which is namedMiniMod, a mini application for seismic modeling. The purpose is to provide a benchmark suite that is, on one hand easy to build and adapt to the state of the art in programming models and changing high performance hardware landscape. On the other hand, the intention is to have a proxy application to actual production geophysical exploration workloads for Oil & Gas exploration, and other geosciences applications based on the wave equation. From top to bottom, we describe the design concepts, algorithms, code structure of the application, and present the benchmark results on different current computer architectures.

cs.DC↗

arXiv2019

Towards an ab initio covariant density functional for nuclear structure

Nuclear structure models built from phenomenological mean fields, the effective nucleon-nucleon interactions (or Lagrangians), and the realistic bare nucleon-nucleon interactions are reviewed. The success of covariant density functional theory (CDFT) to describe nuclear properties and its influence on Brueckner theory within the relativistic framework are focused upon. The challenges and ambiguities of predictions for unstable nuclei without data or for high-density nuclear matter, arising from relativistic density functionals, are discussed. The basic ideas in building an ab initio relativistic density functional for nuclear structure from ab initio calculations with realistic nucleon-nucleon interactions for both nuclear matter and finite nuclei are presented. The current status of fully self-consistent relativistic Brueckner-Hartree-Fock (RBHF) calculations for finite nuclei or neutron drops (ideal systems composed of a finite number of neutrons and confined within an external field) is reviewed. The guidance and perspectives towards an ab initio covariant density functional theory for nuclear structure derived from the RBHF results are provided.

nucl-th↗

arXiv2019

Solving quadratic matrix equations arising in random walks in the quarter plane

Quadratic matrix equations of the kind $A_1X^2+A_0X+A_{-1}=X$ are encountered in the analysis of Quasi--Birth-Death stochastic processes where the solution of interest is the minimal nonnegative solution $G$. In many queueing models, described by random walks in the quarter plane, the coefficients $A_1,A_0,A_{-1}$ are infinite tridiagonal matrices with an almost Toeplitz structure. Here, we analyze some fixed point iterations, including Newton's iteration, for the computation of $G$ and introduce effective algorithms and acceleration strategies which fully exploit the Toeplitz structure of the matrix coefficients and of the current approximation. Moreover, we provide a structured perturbation analysis for the solution $G$. The results of some numerical experiments which demonstrate the effectiveness of our approach are reported.

math.NA↗

arXiv2019

Symmetry energy at supra-saturation densities via the Gravitational Waves from GW170817

Motivated by the historical detection of gravitational waves from GW170817, the neutron star and the neutron drop, i.e., a certain number of neutrons confined in an external field, are systematically investigated by ab initio calculations as well as the nonrelativistic and relativistic state-of-art density functional theories. Strong correlations are found among the neutron star tidal deformability, the neutron star radius, the root-mean-square radii of neutron drops, and the symmetry energies of nuclear matter at supra-saturation densities. From these correlations and the upper limit on the tidal deformability extracted from GW170817, the neutron star radii, the neutron drop radii, and the symmetry energy at twice saturation density are respectively constrained as $R_{1.4M_{\odot}}\leqslant 12.94$ km, $R_{\rm nd} \leqslant 2.36$ fm, and $E_{\mathrm{sym}}(2ρ_0) \leqslant 53.2$ MeV.

nucl-th↗

arXiv2018

Relativistic Brueckner-Hartree-Fock in nuclear matter without the average momentum approximation

Brueckner-Hartree-Fock theory allows to derive the $G$-matrix as an effective interaction between nucleons in the nuclear medium. It depends on the center of mass momentum $\bm{P}$ of the two particles and on the two relative momenta $\bm{q}$ and $\bm{q'}$ before and after the scattering process. In the evaluation of the total energy per particle in nuclear matter usually the angle averaged center of mass momentum approximation has been used. We derive in detail the exact expressions of the angular integrations of the momentum $\bm{P}$ within relativistic Brueckner-Hartree-Fock (RBHF) theory, especially for the case of asymmetric nuclear matter. In order to assess the reliability of the conventional average momentum approximation for the binding energy, the saturation properties of symmetric and asymmetric nuclear matter are systematically investigated based on the realistic Bonn nucleon-nucleon potential. It is found that the exact treatment of the center of mass momentum leads to non-negligible contributions to the higher order physical quantities. The correlation between the symmetry energy $E_{\mathrm{sym}}$, the slope parameter $L$, and the curvature $K_{\mathrm{sym}}$ of the symmetry energy are investigated. The results of our RBHF calculations for the bulk parameters characterizing the equation of state are compared with recent constraints extracted from giant monopole resonance and isospin diffusion experiments.

nucl-th↗

arXiv2018

High precision nuclear mass predictions towards a hundred kilo-electron-volt accuracy

Mass is a fundamental property and an important fingerprint of atomic nucleus. It provides an extremely useful test ground for nuclear models and is crucial to understand energy generation in stars as well as the heavy elements synthesized in stellar explosions. Nuclear physicists have been attempting at developing a precise, reliable, and predictive nuclear model that is suitable for the whole nuclear chart, while this still remains a great challenge even in recent days. Here we employ the Fourier spectral analysis to examine the deviations of nuclear mass predictions to the experimental data and to present a novel way for accurate nuclear mass predictions. In this analysis, we map the mass deviations from the space of nucleon number to its conjugate space of frequency, and are able to pin down the main contributions to the model deficiencies. By using the radial basis function approach we can further isolate and quantify the sources. Taking a pedagogical mass model as an example, we examine explicitly the correlation between nuclear effective interactions and the distributions of mass deviations in the frequency domain. The method presented in this work, therefore, opens up a new way for improving the nuclear mass predictions towards a hundred kilo-electron-volt accuracy, which is argued to be the chaos-related limit for the nuclear mass predictions.

nucl-th↗

arXiv2018

Relativistic Brueckner-Hartree-Fock theory for neutron drops

Neutron drops confined in an external field are studied in the framework of relativistic Brueckner-Hartree-Fock theory using the bare nucleon-nucleon interaction. The ground state energies and radii of neutron drops with even numbers from $N = 4$ to $N=50$ are calculated and compared with results obtained from other nonrelativistic \textit{ab initio} calculations and from relativistic density functional theory. Special attention has been paid to the magic numbers and to the sub-shell closures. The single-particle energies are investigated and the monopole effect of the tensor force on the evolutions of the spin-orbit and the pseudospin-orbit splittings is discussed. The results provide interesting insight of neutron rich systems and can form an important guide for future density functionals.

nucl-th↗

arXiv2018

Spin symmetry in the Dirac sea derived from the bare nucleon-nucleon interaction

The spin symmetry in the Dirac sea has been investigated with relativistic Brueckner-Hartree-Fock theory using the bare nucleon-nucleon interaction. Taking the nucleus $^{16}$O as an example and comparing the theoretical results with the data, the definition of the single-particle potential in the Dirac sea is studied in detail. It is found that if the single-particle states in the Dirac sea are treated as occupied states, the ground state properties are in better agreement with experimental data. Moreover, in this case, the spin symmetry in the Dirac sea is better conserved and it is more consistent with the findings using phenomenological relativistic density functionals.

nucl-th↗

arXiv2018

Effects of Tensor Force in the Relativistic Scheme: A Case Study of Neutron Drops

Tensor force is an important component in the nucleon-nucleon interaction, nevertheless, the role of the tensor force in the spin properties in finite nuclei is much less clear. In this report, we mainly focus on our recent progress on this topic about the relativistic Brueckner-Hartree-Fock theory and the corresponding tensor effects on the spin-orbit splittings in neutron drops. This will serve as an important guide for future microscopic derivations of the relativistic and non-relativistic nuclear energy density functionals.

nucl-th↗

arXiv2018

Effects of tensor forces in nuclear spin-orbit splittings from ab initio calculations

A systematic and specific pattern due to the effects of the tensor forces is found in the evolution of spin-orbit splittings in neutron drops. This result is obtained from relativistic Brueckner-Hartree-Fock theory using the bare nucleon-nucleon interaction. It forms an important guide for future microscopic derivations of relativistic and nonrelativistic nuclear energy density functionals.

nucl-th↗