arXiv subjects
Publications and source records attributed to Jian Liang.
We report an attempt to calculate the deep inelastic scattering structure functions from the hadronic tensor calculated on the lattice. We used the Backus-Gilbert reconstruction method to address the inverse Laplace transformation for the analytic continuation from the Euclidean to the Minkowski space.
We report the results on the proton mass decomposition and also on the related quark and glue momentum fractions. The results are based on overlap valence fermions on four ensembles of $N_f = 2+1$ DWF configurations with three lattice spacings and volumes, and several pion masses including the physical pion mass. With 1-loop perturbative calculation and proper normalization of the glue operator, we find that the $u, d,$ and $s$ quark masses contribute 9(2)\% to the proton mass. The quark energy and glue field energy contribute 31(5)\% and 37(5)\% respectively in the $\overline{MS}$ scheme at $μ= 2$ GeV. The trace anomaly gives the remaining 23(1)\% contribution. The $u,d,s$ and glue momentum fractions in the $\overline{MS}$ scheme are consistent with the global analysis at $μ= 2$ GeV.
In multi-view clustering, different views may have different confidence levels when learning a consensus representation. Existing methods usually address this by assigning distinctive weights to different views. However, due to noisy nature of real-world applications, the confidence levels of samples in the same view may also vary. Thus considering a unified weight for a view may lead to suboptimal solutions. In this paper, we propose a novel localized multi-view subspace clustering model that considers the confidence levels of both views and samples. By assigning weight to each sample under each view properly, we can obtain a robust consensus representation via fusing the noiseless structures among views and samples. We further develop a regularizer on weight parameters based on the convex conjugacy theory, and samples weights are determined in an adaptive manner. An efficient iterative algorithm is developed with a convergence guarantee. Experimental results on four benchmarks demonstrate the correctness and effectiveness of the proposed model.
We present a calculation of the strange and charm quark contributions to the nucleon spin from the anomalous Ward identity (AWI). It is performed with overlap valence quarks on 2+1-flavor domain-wall fermion gauge configurations on a $24^3 \times 64$ lattice with the light sea mass at $m_π = 330$ MeV. To satisfy the AWI, the overlap fermion for the pseudoscalar density and the overlap Dirac operator for the topological density, which do not have multiplicative renormalization, are used to normalize the form factor of the local axial-vector current at finite $q^2$. For the charm quark, we find that the negative pseudoscalar term almost cancels the positive topological term. For the strange quark, the pseudoscalar term is less negative than that of the charm. By imposing the AWI, the strange $g_A(q^2)$ at $q^2 =0$ is obtained by a global fit of the pseudoscalar and the topological form factors, together with $g_A(q^2)$ and the induced pseudoscalar form factor $h_A(q^2)$ at finite $q^2$. The chiral extrapolation to the physical pion mass gives $Δs + Δ{\bar{s}} = -0.0403(44)(78)$.
Object detection when provided image-level labels instead of instance-level labels (i.e., bounding boxes) during training is an important problem in computer vision, since large scale image datasets with instance-level labels are extremely costly to obtain. In this paper, we address this challenging problem by developing an Expectation-Maximization (EM) based object detection method using deep convolutional neural networks (CNNs). Our method is applicable to both the weakly-supervised and semi-supervised settings. Extensive experiments on PASCAL VOC 2007 benchmark show that (1) in the weakly supervised setting, our method provides significant detection performance improvement over current state-of-the-art methods, (2) having access to a small number of strongly (instance-level) annotated images, our method can almost match the performace of the fully supervised Fast RCNN. We share our source code at https://github.com/ZiangYan/EM-WSD.
We report a lattice calculation of nucleon forward matrix elements on a $48^3 \times 96$ lattice at the physical pion mass and a spatial size of 5.5 fm. The $2+1$ flavor dynamical fermion configurations are generated with domain-wall fermions (DWF) and the overlap fermions are adopted for the valence quarks. The isovector $g_A^3$ and $g_S^3$, and the connected insertion part of $g_S^0$ are reported for three source-sink separations. With local current, we obtain $g_A^3 = 1.18(4)$ from a two-state fit. For the quark momentum fraction $\langle x \rangle_{u-d}$, we have included smaller lattices (i.e. $24^3 \times 64$ and $32^3 \times 64$ lattice with pion mass at 330 and 290 MeV respectively) for a fit which includes partially quenched cases as well as finite volume and continuum corrections. A global fit with perturbative renormalization gives $\langle x \rangle_{u-d} (\overline{MS},\, μ= 2\, {\rm GeV}) = 0.170(14)$. We made a cost comparison of calculating the nucleon matrix elements with those from the twisted mass fermion on similar sized lattice at the physical pion point and the domain-wall fermion calculation on the same DWF lattice. We also compare cost with the clover fermion calculation on similar sized lattice at about the same quark mass. The comparison shows that with several improvements, such as many-to-all correlator with grid source and low-mode substitution in the connected insertion and low-mode average in the quark loop can make the overlap as efficient as the twisted-mass and clover fermions in calculating the three-point functions. It is more efficient than the DWF. When the multi-mass feature is invoked, the overlap can be more efficient in reaching the same precision than the single mass comparison made so far.
We report a lattice QCD calculation of the strange quark contribution to the nucleon's magnetic moment and charge radius. This analysis presents the first direct determination of strange electromagnetic form factors including at the physical pion mass. We perform a model-independent extraction of the strange magnetic moment and the strange charge radius from the electromagnetic form factors in the momentum transfer range of $0.051 \,\text{GeV}^2 \lesssim Q^2 \lesssim 1.31 \,\text{GeV}^2 $. The finite lattice spacing and finite volume corrections are included in a global fit with $24$ valence quark masses on four lattices with different lattice spacings, different volumes, and four sea quark masses including one at the physical pion mass. We obtain the strange magnetic moment $G^s_M(0) = - 0.064(14)(09)\, μ_N$. The four-sigma precision in statistics is achieved partly due to low-mode averaging of the quark loop and low-mode substitution to improve the statistics of the nucleon propagator. We also obtain the strange charge radius $\langle r_s^2\rangle_E = -0.0043 (16)(14)\,$ $\text{fm}^2$.
Self-paced learning (SPL) mimics the cognitive mechanism of humans and animals that gradually learns from easy to hard samples. One key issue in SPL is to obtain better weighting strategy that is determined by minimizer function. Existing methods usually pursue this by artificially designing the explicit form of SPL regularizer. In this paper, we focus on the minimizer function, and study a group of new regularizer, named self-paced implicit regularizer that is deduced from robust loss function. Based on the convex conjugacy theory, the minimizer function for self-paced implicit regularizer can be directly learned from the latent loss function, while the analytic form of the regularizer can be even known. A general framework (named SPL-IR) for SPL is developed accordingly. We demonstrate that the learning procedure of SPL-IR is associated with latent robust loss functions, thus can provide some theoretical inspirations for its working mechanism. We further analyze the relation between SPL-IR and half-quadratic optimization. Finally, we implement SPL-IR to both supervised and unsupervised tasks, and experimental results corroborate our ideas and demonstrate the correctness and effectiveness of implicit regularizers.
Lattice QCD calculations with chiral fermions of the $π$N sigma term $σ_{πN}$ and strangeness sigma term $σ_{sN}$ including chiral interpolation with continuum and volume corrections are provided in this work, with the excited-state contaminations subtracted properly. We calculate the scalar matrix element for the light/strange quark directly and find $σ_{πN}=45.9(7.4)(2.8)$ MeV, with the disconnected insertion part contributing 20(12)(4)\%, and $σ_{sN}=40.2(11.7)(3.5)$ MeV, which is somewhat smaller than $σ_{πN}$. The ratio of the strange/light scalar matrix elements is $y$ = 0.09(3)(1).
The $Ω$ baryons with $J^P=3/2^\pm, 1/2^\pm$ are studied on the lattice in the quenched approximation. Their mass levels are ordered as $M_{3/2^+}<M_{3/2^-}\approx M_{1/2^-}<M_{1/2^+}$, as is expected from the constituent quark model. The mass values are also close to those of the four $Ω$ states observed in experiments, respectively. We calculate the Bethe-Salpeter amplitudes of $Ω(3/2^+)$ and $Ω(1/2^+)$ and find there is a radial node for the $Ω(1/2^+)$ Bethe-Salpeter amplitude, which may imply that $Ω(1/2^+)$ is an orbital excitation of $Ω$ baryons as a member of the $(D,L_N^P)=(70,0_2^+)$ supermultiplet in the $SU(6)\bigotimes O(3)$ quark model description. Our results are helpful for identifying the quantum number of experimentally observed $Ω$ states.
The Bethe-Salpeter wave functions of $SU(3)$ pure gauge glueballs are revisited in this study. The ground and the first excited states of scalar and tensor glueballs are identified unambiguously by using the variational method on the basis of large operator sets. We calculate their wave functions in the Coulomb gauge and use two lattices with different lattice spacings to check the discretization artifacts. For ground states, the radial wave functions are approximately Gaussian and the size of the tensor is twice as large as that of the scalar. For the first excited states, the radial nodes are clearly observed for both the scalar and the tensor glueballs, such that they can be interpreted as the first radial excitations. These observations may shed light on the theoretical understanding of the inner structure of glueballs.
In this paper, low-energy scattering of the $(D^{*}\bar{D}^{*})^\pm$ meson system is studied within Lüscher's finite-size formalism using $N_{f}=2$ twisted mass gauge field configurations. With three different pion mass values, the $s$-wave threshold scattering parameters, namely the scattering length $a_0$ and the effective range $r_0$, are extracted in $J^P=1^+$ channel. Our results indicate that, in this particular channel, the interaction between the two vector charmed mesons is weakly repulsive in nature hence do not support the possibility of a shallow bound state for the two mesons, at least for the pion mass values being studied. This study provides some useful information on the nature of the newly discovered resonance-like structure $Z_c(4025)$ observed in various experiments.
In this paper, we will follow Kirchberg's categorical perspective to establish new notions of WEP and QWEP relative to a C$^*$-algebra, and develop similar properties as in the classical WEP and QWEP. Also we will show some examples of relative WEP and QWEP to illustrate the relations with the classical cases. Finally we will apply our notions to recent results on C$^*$-norms.
In the generic domain wall fermion formulation of chiral fermions on the lattice, the zero modes of the four-dimensional Wilson fermion operator with the negative mass parameter $-M_5$ introduce unphysical massive modes propagating in the four-dimensional spacetime. In the free fermion case, the pole mass of this kind of unphysical modes is given by $\tilde{E}=\ln(1-M_5)$, which acquires an imaginary part, $iπ$, when $M_5>1$ and results in an oscillatory behavior of the domain wall fermion propagator in time. The existence of the unphysical modes in the presence of gauge fields is investigated in the mean field approximation, and their physical consequences are discussed. In addition, we also give a semiquantitative criterion for tuning $M_5$ in the realistic numerical study.
In this exploratory lattice study, low-energy scattering of the $(D\bar{D}^{*})^\pm$ meson system is analyzed using lattice QCD with $N_f=2$ twisted mass fermion configurations with three pion mass values. The calculation is performed within single-channel Lüscher's finite-size formalism. The threshold scattering parameters, namely the scattering length $a_0$ and the effective range $r_0$, for the $s$-wave scattering in $J^P=1^+$ channel are extracted. For the cases in our study, the interaction between the two charmed mesons is weakly repulsive. Our lattice results therefore do not support the possibility of a shallow bound state for the two mesons for the pion mass values we studied. This calculation provides some useful information on the nature of the newly discovered resonance-like structure $Z_c(3900)$ by various experimental groups.
The sum of the first n energy levels of the planar Laplacian with constant magnetic field of given total flux is shown to be maximal among triangles for the equilateral triangle, under normalization of the ratio (moment of inertia)/(area)^3 on the domain. The result holds for both Dirichlet and Neumann boundary conditions, with an analogue for Robin (or de Gennes) boundary conditions too. The square similarly maximizes the eigenvalue sum among parallelograms, and the disk maximizes among ellipses. More generally, a domain with rotational symmetry will maximize the magnetic eigenvalue sum among all linear images of that domain. These results are new even for the ground state energy (n=1).