arXiv subjects

Tatsumi Aoyama

Publications and source records attributed to Tatsumi Aoyama.

At least 19 recordsLinked to original sources

arXiv2026

The Peskin-Takeuchi $S$ parameter and vector meson decay constants in $N_f=8$ QCD

We use lattice gauge theory to investigate the Peskin-Takeuchi S parameter in SU(3) gauge theory with eight light fundamental fermions (eight-flavor QCD), a candidate for walking technicolor. We develop a new approach to computing S using the first time moment of current correlators, in addition to the conventional vacuum-polarization method, and obtain consistent results. Finite-volume (FV) corrections are estimated using data at multiple lattice volumes and a formula motivated by chiral perturbation theory. At the smallest fermion mass, the one-doublet contribution is S = 0.281(5)(17), smaller than the value $\sim 0.35$ in scaled-up 2+1-flavor QCD (without subtracting the SM contribution). However, contrary to naive expectations, S toward smaller fermion masses is not necessarily suppressed once FV corrections are included. The resulting Gasser-Leutwyler parameter $L_{10}^r(M_ρ)$ is similar to that in real-world QCD. We also determine the vector and axial-vector meson decay constants, $F_ρ$ and $F_{a_1}$. The ratio $F_ρ/F_π$ is stable against the fermion mass and consistent with the KSRF estimate, implying a $ρππ$ coupling similar to that in real-world QCD. We predict the techni-rho decay width into weak bosons to be $Γ_ρ/M_ρ\lesssim 0.24/N$, where N is the number of weak doublets. We further find that the rho- and $a_1$-dominated Weinberg spectral-function and Das-Mathur-Okubo sum rules are satisfied, yielding S and $L_{10}^r(M_ρ)$ consistent with our main results. We discuss implications for BSM model building.

hep-lat↗

arXiv2026

A novel view of the flavor-singlet spectrum from multi-flavor QCD on the lattice

SU(3) gauge theories with increasing number of light fermions are the templates of strongly interacting sectors and studying their low-energy dynamics and spectrum is important, both for understanding the strong dynamics of QCD itself, but also for discovering viable UV completions of beyond the Standard Model physics. In order to contrast many-flavors strongly interacting theories with QCD on a quantitative footing, we use Lattice Field Theory simulations. We focus on the study of the flavor-singlet spectrum in the scalar and pseudoscalar channels: this is an interesting probe of the dynamics of the strongly interacting sector, as reminded by the QCD case with the $f_0(500)$ ($σ$) and $η^\prime$ mesons. The hierarchy of the spectrum of a strongly coupled new gauge sector of the Standard Model defines the potential reach of future colliders for new physics discoveries. In addition to a novel hierarchy with light scalars, introducing many light flavors at fixed number of colors can influence the dynamics of the lightest flavor-singlet pseudoscalar. We present a complete lattice study of both these flavor-singlet channels on high-statistics gauge ensembles generated by the LatKMI collaboration with 4, 8, and 12 copies of light mass-degenerate fermions. We also present other hadron masses on the lightest ensemble for $N_f=8$ generated by the LatKMI collaboration and discuss the chiral extrapolation of the spectrum in this particular theory. We contrast the results to $N_f=4$ simulations and previous results of $N_f=12$ simulations.

hep-lat↗

arXiv2026

Update of PHYSBO: Improving Usability and Portability of Bayesian Optimization for Physics and Materials Research

Bayesian optimization (BO) is widely used to accelerate physics and materials research, where objective function evaluations are computationally or experimentally expensive. While many BO frameworks focus on algorithmic efficiency, practical usability and portability are equally critical for sustained use in real research environments. PHYSBO is a Bayesian optimization library designed to address these needs by enabling optimization over user-defined candidate pools and by supporting domain-specific problem settings. This paper presents the major updates introduced in PHYSBO versions 2 and 3, with a focus on improvements in usability, portability, and practical deployment rather than on new optimization algorithms. In PHYSBO version 2, the software license was changed from GPL to MPL to improve compatibility with a wider range of research and software ecosystems. Building on this revision, PHYSBO version 3 introduces a set of implementation-oriented updates aimed at improving usability and portability, without modifying the core optimization algorithms. These updates include improvements in computational performance and scalability, extended support for multi-objective optimization, the introduction of range-based policies for continuous-variable optimization, the removal of environment-dependent components such as tightly coupled Cython modules, and compatibility with NumPy 2. These improvements reduce the technical and organizational burden on users, enabling PHYSBO to be deployed across diverse computing environments and research workflows. By emphasizing portability and ease of integration while maintaining sufficient performance, PHYSBO version 3 is positioned as a sustainable research infrastructure for Bayesian optimization in physics and materials science.

physics.comp-ph↗

arXiv2026

Mixed precision solvers with half-precision floating point numbers for Lattice QCD on A64FX processor

We investigate the use of half-precision floating-point numbers (FP16) in mixed-precision linear solvers for lattice QCD simulations. Since the emergence of GPUs for general-purpose, mixed-precision algorithms that combine single-precision (FP32) with double-precision (FP64) arithmetics have become widely used in this field and others. While FP32-based methods are now well established, we examine the practicality of using FP16. In this work, we introduce rescaling steps in both the outer iterative refinement step and the inner BiCGStab solver to avoid numerical instability. In our experiments with a simple Wilson kernel, the solver shows improved stability, and the additional iteration count compared to the FP64 version remains within 20\%, indicating that the FP16 version is practical for use. We believe that the proposed rescaling methods can also benefit other mixed precision preconditioners in avoiding underflows.

hep-lat↗

arXiv2025

Project For Advancement of Software Usability in Materials Science

The Institute for Solid State Physics (ISSP) at The University of Tokyo has been carrying out a software development project named ``the Project for Advancement of Software Usability in Materials Science (PASUMS)". Since the launch of PASUMS, various open-source software programs have been developed/advanced, including ab initio calculations, effective model solvers, and software for machine learning. We also focus on activities that make the software easier to use, such as developing comprehensive computing tools that enable efficient use of supercomputers and interoperability between different software programs. We hope to contribute broadly to developing the computational materials science community through these activities.

cs.SE↗

arXiv2025

Verification of the tenth-Order QED contribution to the anomalous magnetic moment of the electron from diagrams without fermion loops

A discrepancy of approximately 5$σ$ exists between the two known results for the tenth-order QED contribution to the anomalous magnetic moment of the electron, calculated from Feynman vertex diagrams without fermion loops. To investigate this, we decomposed this contribution into 389 parts based on a self-energy diagram representation, enabling a diagram-by-diagram numerical comparison of the two calculations. No significant discrepancies were found for individual diagrams. However, the numerical differences of the 98 diagrams sharing a common structure were not randomly distributed. The accumulation of these differences resulted in the 5$σ$ discrepancy. A recalculation with increased statistics in the Monte Carlo integration was performed for these 98 diagrams. By replacing the old values with the new ones for these 98 integrals, we have obtained a revised result of $6.800 \pm 0.128$, thereby resolving the discrepancy.

hep-ph↗

arXiv2025

TeNeS-v2: Enhancement for Real-Time and Finite Temperature Simulations of Quantum Many-Body Systems

Quantum many-body systems are challenging targets for computational physics due to their large degrees of freedom. The tensor networks, particularly Tensor Product States (TPS) and Projected Entangled Pair States (PEPS), effectively represent these systems on two-dimensional lattices. However, the technical complexity of TPS/PEPS-based coding is often too much for researchers to handle. To reduce this problem, we developed TeNeS (Tensor Network Solver). This paper introduces TeNeS-v2, which extends TeNeS with real-time and finite temperature simulations, providing deeper insights into quantum many-body systems. We detail the new algorithms, input/output design, and application examples, demonstrating TeNeS-v2's applicability to various quantum spin and Bose models on two-dimensional lattices.

cond-mat.str-el↗

arXiv2024

Scale setting and hadronic properties in light quark sector with $(2+1)$-flavor Wilson fermions at the physical point

We report scale setting and hadronic properties for our new lattice QCD gauge configuration set (HAL-conf-2023). We employ $(2+1)$-flavor nonperturbatively improved Wilson fermions with stout smearing and the Iwasaki gauge action on a $96^4$ lattice, and generate configurations of 8,000 trajectories at the physical point. We show the basic properties of the configurations such as the plaquette value, topological charge distribution and their auto-correlation times. The scale setting is performed by detailed analyses of the $Ω$ baryon mass. We calculate the physical results of quark masses, decay constants of pseudoscalar mesons and single hadron spectra in light quark sector. The masses of the stable hadrons are found to agree with the experimental values within a sub-percent level.

hep-lat↗

arXiv2024

H-wave -- A Python package for the Hartree-Fock approximation and the random phase approximation

H-wave is an open-source software package for performing the Hartree--Fock approximation (HFA) and random phase approximation (RPA) for a wide range of Hamiltonians of interacting fermionic systems. In HFA calculations, H-wave examines the stability of several symmetry-broken phases, such as anti-ferromagnetic and charge-ordered phases, in the given Hamiltonians at zero and finite temperatures. Furthermore, H-wave calculates the dynamical susceptibilities using RPA to examine the instability toward the symmetry-broken phases. By preparing a simple input file for specifying the Hamiltonians, users can perform HFA and RPA for standard Hamiltonians in condensed matter physics, such as the Hubbard model and its extensions. Additionally, users can use a Wannier90-like format to specify fermionic Hamiltonians. A Wannier90 format is implemented in RESPACK to derive ab initio Hamiltonians for solids. HFA and RPA for the ab initio Hamiltonians can be easily performed using H-wave. In this paper, we first explain the basis of HFA and RPA, and the basic usage of H-wave, including download and installation. Thereafter, the input file formats implemented in H-wave, including the Wannier90-like format for specifying the interacting fermionic Hamiltonians, are discussed. Finally, we present several examples of H-wave such as zero-temperature HFA calculations for the extended Hubbard model on a square lattice, finite-temperature HFA calculations for the Hubbard model on a cubic lattice, and RPA in the extended Hubbard model on a square lattice.

cond-mat.str-el↗

arXiv2023

Configuration sampling in multi-component multi-sublattice systems enabled by ab Initio Configuration Sampling Toolkit (abICS)

Simulation of the intermediate levels of disorder found in multi-component multi-sublattice systems in various functional materials is a challenging issue, even for state-of-the-art methodologies based on first-principles calculation. Here, we introduce our open-source package ab Initio Configuration Sampling Toolkit (abICS), which combines high-throughput first-principles calculations, machine learning, and parallel extended ensemble sampling in an active learning setting to enable such simulations. The theoretical background is reviewed in some detail followed by brief notes on usage of the software. In addition, our recent applications of abICS to multi-component ionic systems and their interfaces for energy applications are reviewed as demonstration of the power of this approach.

cond-mat.mtrl-sci↗

arXiv2023

Bridge++ 2.0: Benchmark results on supercomputer Fugaku

Bridge++ is a general-purpose code set for lattice QCD simulations aiming at a readable, extensible, and portable code while keeping practical high performance. The new version 2.0 employs machine-dependent optimization, enabling flexible data layout in float/double precision, while it was fixed layout and only with the double precision in previous versions. We report the performance on supercomputer Fugaku with Arm A64FX-SVE architecture by Fujitsu.

hep-lat↗

arXiv2022

Nuclear force with LapH smearing

The nuclear forces are determined by combining the HAL QCD method and a new type of source smearing technique. The new smearing is a projection to a space spanned by the lowest-lying eigenvectors of the free Laplacian operator on a lattice, which enables efficient calculation of hadron correlators at an affordable cost by utilizing the hadron-level momentum conservations. We find that this new approach reduces the statistical and systematic errors in the resultant nuclear forces.

hep-lat↗

arXiv2021

General purpose lattice QCD code set Bridge++ 2.0 for high performance computing

Bridge++ is a general-purpose code set for a numerical simulation of lattice QCD aiming at a readable, extensible, and portable code while keeping practically high performance. The previous version of Bridge++ is implemented in double precision with a fixed data layout. To exploit the high arithmetic capability of new processor architecture, we extend the Bridge++ code so that optimized code is available as a new branch, i.e., an alternative to the original code. This paper explains our strategy of implementation and displays application examples to the following architectures and systems: Intel AVX-512 on Xeon Phi Knights Landing, Arm A64FX-SVE on Fujitsu A64FX (Fugaku), NEC SX-Aurora TSUBASA, and GPU cluster with NVIDIA V100.

hep-lat↗

arXiv2020

Partial wave decomposition on the lattice and its applications to the HAL QCD method

The approximated partial wave decomposition method to the discrete data on a cubic lattice, developed by C. W. Misner, is applied to the calculation of $S$-wave hadron-hadron scatterings by the HAL QCD method in lattice QCD. We consider the Nambu-Bethe-Salpeter (NBS) wave function for the spin-singlet $Λ_c N$ system calculated in the $(2+1)$-flavor QCD on a $(32a~\mathrm{fm})^3$ lattice at the lattice spacing $a\simeq0.0907$ fm and $m_π\simeq 700$ MeV. We find that the $l=0$ component can be successfully extracted by Misner's method from the NBS wave function projected to $A_1^+$ representation of the cubic group, which contains small $l\ge 4$ components. Furthermore, while the higher partial wave components are enhanced so as to produce significant comb-like structures in the conventional HAL QCD potential if the Laplacian approximated by the usual second order difference is applied to the NBS wave function, such structures are found to be absent in the potential extracted by Misner's method, where the Laplacian can be evaluated analytically for each partial wave component. Despite the difference in the potentials, two methods give almost identical results on the central values and on the magnitude of statistical errors for the fits of the potentials, and consequently on the scattering phase shifts. This indicates not only that Misner's method works well in lattice QCD with the HAL QCD method but also that the contaminations from higher partial waves in the study of $S$-wave scatterings are well under control even in the conventional HAL QCD method. It will be of interest to study interactions in higher partial wave channels in the HAL QCD method with Misner's decomposition, where the utility of this new technique may become clearer.

hep-lat↗

arXiv2020

The HAL QCD potential in $I=1$ $ππ$ system with the $ρ$ meson bound state

In this paper, we investigate the HAL QCD potential in the $I=1$ $ππ$ scattering using the hybrid method for all-to-all propagators, in which a propagator is approximated by low-eigenmodes and the remaining high-eigenmode part is stochastically estimated. To verify the applicability of the hybrid method to systems containing quark creation$/$annihilation contributions such as the $ρ$ meson, we calculate the $I=1$ $ππ$ potential with the 2+1 flavor gauge configurations on $16^3 \times 32$ lattice at the lattice spacing $a \approx 0.12$ fm and $(m_π,m_ρ) \approx (870, 1230)$ MeV, in which the $ρ$ meson appears as a deeply-bound state. While we find that the naive stochastic evaluations for quark creation$/$annihilation contributions lead to extremely large statistical fluctuations, additional noise reduction methods enable us to obtain a sufficiently precise potential, which shows a strong attractive force. We also confirm that the binding energy and $k^3 \cot δ$ obtained from our potential are roughly consistent with an existing $ρ$ meson bound state, within a large systematic error associated with our calculation, whose possible origin is also discussed.

hep-lat↗

arXiv2019

Study of the pion-pion scatterings with a combination of all-to-all propagators and the HAL QCD method

In this paper, we report recent developments of the HAL QCD method for two hadron systems which contain quark annihilation processes using all-to-all quark propagators. We employ the hybrid method for all-to-all propagators, which combines a low-mode spectral decomposition of the quark propagator and stochastic estimators for remaining high modes, to evaluate the HAL QCD potentials for the first time. Using this method, we investigate the $I= 1,2$ $ππ$ scatterings at $m_π \approx 870$ MeV. In the $I=2$ study, we study how statistical fluctuations of the HAL QCD potentials are increased due to stochastic estimators in the hybrid method, compared with the conventional one without them. We find that we can reduce statistical fluctuations by dilutions of stochastic noises in order to obtain sufficiently precise results, which turn out to be consistent with conventional results without all-to-all propagators. In the $I=1$ $ππ$ case, which contains quark annihilation processes, we find that statistical fluctuations are further enhanced due to noise contaminations in annihilation processes. We, however, confirm that we can also reduce such statistical fluctuations to obtain the potential with a reasonable precision as long as we further increase a degree of dilutions at a price of large numerical costs and take an appropriate scheme for the potential.

hep-lat↗

arXiv2019

$I=2$ $ππ$ potential in the HAL QCD method with all-to-all propagators

In this paper, we perform the first application of the hybrid method (exact low modes plus stochastically estimated high modes) for all-to-all propagators to the HAL QCD method. We calculate the HAL QCD potentials in the $I=2$ $ππ$ scattering in order to see how statistical fluctuations of the potential behave under the hybrid method. All of the calculations are performed with the 2+1 flavor gauge configurations on $16^3 \times 32$ lattice at the lattice spacing $a \approx 0.12$ fm and $m_π \approx 870$ MeV. It is revealed that statistical errors for the potential are enhanced by stochastic noises introduced by the hybrid method, which, however, are shown to be reduced by increasing the level of dilutions, in particular, that of space dilutions. From systematic studies, we obtain a guiding principle for a choice of dilution types/levels and a number of eigenvectors to reduce noise contaminations to the potential while keeping numerical costs reasonable. We also confirm that we can obtain the scattering phase shifts for the $I=2$ $ππ$ system by the hybrid method within a reasonable numerical cost, which are consistent with the result obtained with the conventional method. The knowledge we obtain in this study will become useful to investigate hadron resonances which require quark annihilation diagrams such as the $ρ$ meson by the HAL QCD potential with the hybrid method.

hep-lat↗

arXiv2018

Revised and Improved Value of the QED Tenth-Order Electron Anomalous Magnetic Moment

In order to improve the theoretical prediction of the electron anomalous magnetic moment $a_e$ we have carried out a new numerical evaluation of the 389 integrals of Set V, which represent 6,354 Feynman vertex diagrams without lepton loops. During this work, we found that one of the integrals, called $X024$, was given a wrong value in the previous calculation due to an incorrect assignment of integration variables. The correction of this error causes a shift of $-1.25$ to the Set~V contribution, and hence to the tenth-order universal (i.e., mass-independent) term $ A_1^{(10)}$. The previous evaluation of all other 388 integrals is free from errors and consistent with the new evaluation. Combining the new and the old (excluding $X024$) calculations statistically, we obtain $7.606~(192) (α/π)^5$ as the best estimate of the Set V contribution. Including the contribution of the diagrams with fermion loops, the improved tenth-order universal term becomes $A_1^{(10)}=6.678~(192)$. Adding hadronic and electroweak contributions leads to the theoretical prediction $a_e (\text{theory}) =1~159~652~182.032~(720)\times 10^{-12}$. From this and the best measurement of $a_e$, we obtain the inverse fine-structure constant $α^{-1}(a_e) = 137.035~999~1491~(331)$. The theoretical prediction of the muon anomalous magnetic moment is also affected by the update of QED contribution and the new value of $α$, but the shift is much smaller than the theoretical uncertainty.

hep-ph↗