T. Izubuchi

Brookhaven National Laboratory, New York, New York, United States

Are you T. Izubuchi?

Claim your profile

Publications (130)325.69 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We report the first lattice QCD calculation of the hadronic vacuum polarization disconnected contribution to the muon anomalous magnetic moment at physical pion mass. The calculation uses a refined noise-reduction technique which enabled the control of statistical uncertainties at the desired level with modest computational effort. Measurements were performed on the $48^3 \times 96$ physical-pion-mass lattice generated by the RBC and UKQCD collaborations. We find $a_\mu^{\rm HVP~(LO)~DISC} = -9.6(3.3)(2.3)\times 10^{-10}$, where the first error is statistical and the second systematic.
    Full-text · Article · Dec 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: The application of the Dirac operator on a spinor field, the Dslash operation, is the most computation-intensive part of the lattice QCD simulations. It is often the key kernel to optimize to achieve maximum performance on various platforms. Here we report on a project to optimize the domain wall fermion Dirac operator in Columbia Physics System (CPS) using the R-Stream source-to-source compiler. Our initial target platform is the Intel PC clusters. We discuss the optimization strategies involved before and after the automatic code generation with R-Stream and present some preliminary benchmark results.
    No preview · Article · Dec 2015
  • Source
    Eigo Shintani · Thomas Blum · Taku Izubuchi · Amarjit Soni
    [Show abstract] [Hide abstract]
    ABSTRACT: We present a lattice calculation of the neutron and proton electric dipole moments (EDM's) with $N_f=2+1$ flavors of domain-wall fermions. The neutron and proton EDM form factors are extracted from three-point functions at the next-to-leading order in the $\theta$ vacuum of QCD. In this computation, we use pion masses 0.33 and 0.42 GeV and 2.7 fm$^3$ lattices with Iwasaki gauge action and a 0.17 GeV pion and 4.6 fm$^3$ lattice with I-DSDR gauge action, all generated by the RBC and UKQCD collaborations. The all-mode-averaging technique enables an efficient and high statistics calculation. Chiral behavior of lattice EDM's is discussed in the context of baryon chiral perturbation theory. In addition, we also show numerical evidence on relationship of three- and two-point correlation function with local topological distribution.
    Preview · Article · Dec 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: We report the first lattice QCD calculation of the complex kaon decay amplitude A_{0} with physical kinematics, using a 32^{3}×64 lattice volume and a single lattice spacing a, with 1/a=1.3784(68) GeV. We find Re(A_{0})=4.66(1.00)(1.26)×10^{-7} GeV and Im(A_{0})=-1.90(1.23)(1.08)×10^{-11} GeV, where the first error is statistical and the second systematic. The first value is in approximate agreement with the experimental result: Re(A_{0})=3.3201(18)×10^{-7} GeV, while the second can be used to compute the direct CP-violating ratio Re(ϵ^{'}/ϵ)=1.38(5.15)(4.59)×10^{-4}, which is 2.1σ below the experimental value 16.6(2.3)×10^{-4}. The real part of A_{0} is CP conserving and serves as a test of our method while the result for Re(ϵ^{'}/ϵ) provides a new test of the standard model theory of CP violation, one which can be made more accurate with increasing computer capability.
    No preview · Article · Nov 2015 · Physical Review Letters
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The current measurement of muonic $g - 2$ disagrees with the theoretical calculation by about 3 standard deviations. Hadronic vacuum polarization (HVP) and hadronic light by light (HLbL) are the two types of processes that contribute most to the theoretical uncertainty. The current value for HLbL is still given by models. I will describe results from a first-principles lattice calculation with a 139 MeV pion in a box of 5.5 fm extent. Our current numerical strategies, including noise reduction techniques, evaluating the HLbL amplitude at zero external momentum transfer, and important remaining challenges, in particular those associated with finite volume effects, will be discussed.
    Preview · Article · Nov 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: For reliable comparison of the standard model prediction to the muon g-2 with its experimental value, the hadronic light-by-light scattering (HLbL) contribution must be calculated by lattice QCD simulation. HLbL contribution has many types of disconnected-type diagrams. Here, we start with recalling the point that must be taken care of in every method to calculate them by lattice QCD, and present one concrete method called nonperturbative QED method.
    Preview · Article · Nov 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The anomalous magnetic moment of muon, $g-2$, is a very precisely measured quantity. However, the current measurement disagrees with standard model by about 3 standard deviations. Hadronic vacuum polarization and hadronic light by light are the two types of processes that contribute most to the theoretical uncertainty. I will describe how lattice methods are well-suited to provide a first-principle's result for the hadronic light by light contribution, the various numerical strategies that are presently being used to evaluate it, our current results and the important remaining challenges which must be overcome.
    Preview · Article · Sep 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Existing high-level, source-to-source compilers can accept input programs in a high-level language (e.g., C) and perform complex automatic parallelization and other mappings using various optimizations. These optimizations often require trade-offs and can benefit from the user's involvement in the process. However, because of the inherent complexity, the barrier to entry for new users of these high-level optimizing compilers can often be high. We propose visualization as an effective gateway for non-expert users to gain insight into the effects of parameter choices and so aid them in the selection of levels best suited to their specific optimization goals. A popular optimization paradigm is polyhedral mapping which achieves optimization by loop transformations. We have augmented a commercial polyhedral-model source-to-source compiler (R-Stream) with an interactive visual tool we call the Polyhedral User Mapping and Assistant Visualizer (PUMA-V). PUMA-V is tightly integrated with the R-Stream source-to-source compiler and allows users to explore the effects of difficult mappings and express their goals to optimize trade-offs. It implements advanced multivariate visualization paradigms such as parallel coordinates and correlation graphs and applies them in the novel setting of compiler optimizations. We believe that our tool allows programmers to better understand complex program transformations and deviations of mapping properties on well understood programs. This in turn will achieve experience and performance portability across programs architectures as well as expose new communities in the computational sciences to the rich features of auto-parallelizing high-level source-to-source compilers.
    Full-text · Conference Paper · Sep 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We report the first lattice QCD calculation of the complex kaon decay amplitude $A_0$ with physical kinematics, using a $32^3\times 64$ lattice volume and a single lattice spacing $a$, with $1/a= 1.379(7)$ GeV. We find Re$(A_0) = 4.62(0.99)(1.20) \times 10^{-7}$ GeV and Im$(A_0) = -1.90(1.22)(1.04) \times 10^{-11}$ GeV, where the first error is statistical and the second systematic. The first value is in approximate agreement with the experimental result: Re$(A_0) = 3.3201(18) \times 10^{-7}$ GeV while the second can be used to compute the direct CP violating ratio Re$(\varepsilon'/\varepsilon)=1.36(5.21)(4.49)\times 10^{-4}$, which is $2.1\sigma$ below the experimental value $16.6(2.3)\times10^{-4}$. The real part of $A_0$ is CP conserving and serves as a test of our method while the result for Re$(\varepsilon'/\varepsilon)$ provides a new test of the standard-model theory of CP violation, one which can be made more accurate with increasing computer capability.
    Full-text · Article · May 2015
  • Source
    Christoph Lehner · Taku Izubuchi
    [Show abstract] [Hide abstract]
    ABSTRACT: We present a method to couple finite-volume QCD to infinite-volume QED by an appropriate twist-averaging procedure. We demonstrate the prescription numerically for the leading-order hadronic contribution to the anomalous magnetic moment of the muon and the electro-magnetic pion mass splitting.
    Preview · Article · Mar 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We calculate the form factors for B->pi l nu & Bs->K l nu decay in lattice QCD. We use the (2+1)-flavor RBC-UKQCD gauge field-ensembles generated with the domain-wall fermion and Iwasaki gauge actions. For the b quarks we use the anisotropic clover action with the relativistic heavy-quark interpretation. We analyze data at 2 lattice spacings a~0.11, 0.086 fm with pion masses as light as M_pi~290 MeV. We extrapolate our numerical results to the physical light-quark masses and to the continuum and interpolate in the pion/kaon energy using SU(2) "hard-pion" chiral perturbation theory. We provide complete systematic error budgets for the vector & scalar form factors f+(q^2) & f0(q2) for B->pi l nu & Bs ->K l nu at 3 momenta that span the q^2 range accessible in our numerical simulations. Next we extrapolate these results to q^2 = 0 using a model-independent z-parameterization based on analyticity & unitarity. We present our final results for f+(q^2) & f0(q^2) as the z coefficients and matrix of correlations between them; this parameterizes the form factors over the entire kinematic range. Our results agree with other 3-flavor lattice-QCD determinations using staggered light quarks, and have comparable precision, thereby providing important independent checks. Both B->pi l nu & Bs->K l nu decays enable determinations of the CKM element |Vub|. To illustrate this, we perform a combined z-fit of our numerical B ->pi l nu form-factor data with experimental measurements of the branching fraction; we obtain |Vub| = 3.61(32) x 10^{-3}, where the error includes statistical and systematic uncertainties. The same approach can be applied to Bs->K l nu to provide an alternative determination of |Vub| once the process has been measured experimentally. Finally, we make predictions for B->pi l nu & Bs->K l nu differential branching fractions and forward-backward asymmetries in the Standard Model.
    Preview · Article · Jan 2015 · Physical Review D
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We report initial nucleon structure results computed on lattices with 2+1 dynamical M\"obius domain wall fermions at the physical point generated by the RBC and UKQCD collaborations. At this stage, we evaluate only connected quark contributions. In particular, we discuss the nucleon vector and axial-vector form factors, nucleon axial charge and the isovector quark momentum fraction. From currently available statistics, we estimate the stochastic accuracy of the determination of $g_A$ and $ _{u-d}$ to be around 10%, and we expect to reduce that to 5% within the next year. To reduce the computational cost of our calculations, we extensively use acceleration techniques such as low-eigenmode deflation and all-mode-averaging (AMA). We present a method for choosing optimal AMA parameters.
    Full-text · Article · Dec 2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We present results for several light hadronic quantities ($f_\pi$, $f_K$, $B_K$, $m_{ud}$, $m_s$, $t_0^{1/2}$, $w_0$) obtained from simulations of 2+1 flavor domain wall lattice QCD with large physical volumes and nearly-physical pion masses at two lattice spacings. We perform a short, O(3)%, extrapolation in pion mass to the physical values by combining our new data in a simultaneous chiral/continuum `global fit' with a number of other ensembles with heavier pion masses. We use the physical values of $m_\pi$, $m_K$ and $m_\Omega$ to determine the two quark masses and the scale - all other quantities are outputs from our simulations. We obtain results with sub-percent statistical errors and negligible chiral and finite-volume systematics for these light hadronic quantities, including: $f_\pi$ = 130.2(9) MeV; $f_K$ = 155.5(8) MeV; the average up/down quark mass and strange quark mass in the $\overline {\rm MS}$ scheme at 3 GeV, 2.997(49) and 81.64(1.17) MeV respectively; and the neutral kaon mixing parameter, $B_K$, in the RGI scheme, 0.750(15) and the $\overline{\rm MS}$ scheme at 3 GeV, 0.530(11).
    Full-text · Article · Nov 2014
  • Z Bai · N H Christ · T Izubuchi · C T Sachrajda · A Soni · J Yu
    [Show abstract] [Hide abstract]
    ABSTRACT: We report on the first complete calculation of the K_{L}-K_{S} mass difference, ΔM_{K}, using lattice QCD. The calculation is performed on a 2+1 flavor, domain wall fermion ensemble with a 330 MeV pion mass and a 575 MeV kaon mass. We use a quenched charm quark with a 949 MeV mass to implement Glashow-Iliopoulos-Maiani cancellation. For these heavier-than-physical particle masses, we obtain ΔM_{K}=3.19(41)(96)×10^{-12} MeV, quite similar to the experimental value. Here the first error is statistical, and the second is an estimate of the systematic discretization error. An interesting aspect of this calculation is the importance of the disconnected diagrams, a dramatic failure of the Okubo-Zweig-Iizuka rule.
    No preview · Article · Sep 2014 · Physical Review Letters
  • [Show abstract] [Hide abstract]
    ABSTRACT: The composition of nucleons has long been known to be sub-atomic particles called quarks and gluons, which interact through the strong force and theoretically can be described by Quantum Chromodynamics (QCD). Lattice QCD (LQCD), in which the continuous space-time is translated into grid points on a four-dimensional lattice and ab initio Monte Carlo simulations are performed, is by far the only model-independent method to study QCD with controllable errors. We report the successful application of a novel algorithm, All-Mode-Averaging, in the LQCD calculations of nucleon internal structure on the Gordon supercomputer our award of roughly 6 million service units through XSEDE. The application of AMA resulted in as much as a factor of 30 speedup in computational efficiency.
    No preview · Article · Jul 2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The form factor that yields the light-by-light scattering contribution to the muon anomalous magnetic moment is computed in lattice QCD+QED and QED. A non-perturbative treatment of QED is used and is checked against perturbation theory. The hadronic contribution is calculated for unphysical quark and muon masses, and only the diagram with a single quark loop is computed. Statistically significant signals are obtained. Initial results appear promising, and the prospect for a complete calculation with physical masses and controlled errors is discussed.
    Full-text · Article · Jul 2014 · Physical Review Letters
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Neutral $B$ meson mixing matrix elements and $B$ meson decay constants are calculated. Static approximation is used for $b$ quark and domain-wall fermion formalism is employed for light quarks. The calculations are carried out on $2+1$ flavor dynamical ensembles generated by RBC/UKQCD Collaborations with lattice spacings $0.086$fm ($a^{-1}\sim 2.3$GeV) and $0.11$fm ($1.7$GeV), and a fixed physical spatial volume of about $(2.7{\rm fm})^3$. In the static quark action, link-smearings are used to improve the signal-to-noise ratio. We employ two kinds of link-smearings, HYP1 and HYP2, and their results are combined in taking the continuum limit. For the matching between the lattice and the continuum theory, one-loop perturbative $O(a)$ improvements are made to reduce discretization errors. As the most important quantity of this work, we obtain SU(3) breaking ratio $\xi=1.208(60)$, where the error includes statistical and systematic one. We also find other neutral $B$ meson mixing quantities $f_B\sqrt{\hat{B}_B}=240(22)$MeV, $f_{B_s}\sqrt{\hat{B}_{B_s}}=290(22)$MeV, $\hat{B}_B=1.17(22)$, $\hat{B}_{B_s}=1.22(13)$ and $B_{B_s}/B_B=1.028(74)$, $B$ meson decay constants $f_B=219(17)$MeV, $f_{B_s}=264(19)$MeV and $f_{B_s}/f_B=1.193(41)$, in the static limit of $b$ quark.
    Preview · Article · Jun 2014 · Physical Review D
  • Source
    Z. Bai · N. H. Christ · T. Izubuchi · C. T. Sachrajda · A. Soni · J. Yu
    [Show abstract] [Hide abstract]
    ABSTRACT: We report on the first complete calculation of the $K_L-K_S$ mass difference, $\Delta M_K$, using lattice QCD. The calculation is performed on a 2+1 flavor, domain wall fermion (DWF) ensemble with a 330~MeV pion mass and a 575~MeV kaon mass. We use a quenched charm quark with a 949~MeV mass to implement Glashow-Iliopoulos-Maiani (GIM) cancellation. For these heavier-than-physical particle masses, we obtain $\Delta M_K =3.19(41)(96)\times 10^{-12}$~MeV, quite similar to the experimental value. Here the first error is statistical and the second is an estimate of the systematic discretization error. An interesting aspect of this calculation is the importance of the disconnected diagrams, a dramatic failure of the OZI rule.
    Full-text · Article · Jun 2014 · Physical Review Letters
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We calculate the B-meson decay constants f_B, f_Bs, and their ratio in unquenched lattice QCD using domain-wall light quarks and relativistic b-quarks. We use gauge-field ensembles generated by the RBC and UKQCD collaborations using the domain-wall fermion action and Iwasaki gauge action with three flavors of light dynamical quarks. We analyze data at two lattice spacings of a ~ 0.11, 0.086 fm with unitary pion masses as light as M_pi ~ 290 MeV; this enables us to control the extrapolation to the physical light-quark masses and continuum. For the b-quarks we use the anisotropic clover action with the relativistic heavy-quark interpretation, such that discretization errors from the heavy-quark action are of the same size as from the light-quark sector. We renormalize the lattice heavy-light axial-vector current using a mostly nonperturbative method in which we compute the bulk of the matching factor nonperturbatively, with a small correction, that is close to unity, in lattice perturbation theory. We also improve the lattice heavy-light current through O(alpha_s a). We extrapolate our results to the physical light-quark masses and continuum using SU(2) heavy-meson chiral perturbation theory, and provide a complete systematic error budget. We obtain f_B0 = 196.2(15.7) MeV, f_B+ = 195.4(15.8) MeV, f_Bs = 235.4(12.2) MeV, f_Bs/f_B0 = 1.193(59), and f_Bs/f_B+ = 1.220(82), where the errors are statistical and total systematic added in quadrature. These results are in good agreement with other published results and provide an important independent cross check of other three-flavor determinations of B-meson decay constants using staggered light quarks.
    Preview · Article · Apr 2014 · Physical Review D
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The real and imaginary parts of the $K_L-K_S$ mixing matrix receive contributions from all three charge-2/3 quarks: up, charm and top. These give both short- and long-distance contributions which are accessible through a combination of perturbative and lattice methods. We will discuss a strategy to compute both the mass difference, $\Delta M_K$ and $\epsilon_K$ to sub-percent accuracy, looking in detail at the contributions from each of the three CKM matrix element products $V_{id}^*V_{is}$ for $i=u, c$ and $t$ as described in Ref. [1]
    Full-text · Article · Feb 2014

Publication Stats

3k Citations
325.69 Total Impact Points

Institutions

  • 2002-2015
    • Brookhaven National Laboratory
      • Physics Department
      New York, New York, United States
  • 2012
    • University of Connecticut
      • Department of Physics
      Storrs, CT, United States
  • 2000-2009
    • Kanazawa University
      • • Institute for Theoretical Physics
      • • Department of Physics
      Kanazawa, Ishikawa, Japan
  • 2005-2006
    • RIKEN
      • Theoretical Nuclear Physics Laboratory
      Вако, Saitama, Japan
  • 1997-2000
    • University of Tsukuba
      • Centre for Computational Sciences
      Tsukuba, Ibaraki, Japan
  • 1995-1996
    • The University of Tokyo
      • Department of Physics
      白山, Tōkyō, Japan