arXiv subjects

Yushan Su

Publications and source records attributed to Yushan Su.

30 records · Page 2Linked to original sources

arXiv2023

Precision Control in Lattice Calculation of $x$-dependent Pion Distribution Amplitude

We present a new Bjorken $x$-dependence analysis of a previous lattice quantum chromodynamics data for the pion distribution amplitude from MILC configurations with three lattice spacing $a=0.06,0.09, 0.12$~fm. A leading renormalon resummation in renormalization as well as the perturbative matching kernel in the framework of large momentum expansion generates the power accuracy of the matching to the light-cone amplitude. Meanwhile, a small momentum log resummation is implemented for both the quark momentum $xP_z$ and the antiquark momentum $(1-x)P_z$ inside a meson of boost momentum $P_z$ up to 1.72 GeV along the $z$ direction, allowing us to have more accurate determination of the $x$-dependence in the middle range. Finally, we use the complementarity between the short-distance factorization and the large momentum expansion to constrain the endpoint regions $x\sim 0, 1$, thus obtaining the full-range $x$-dependence of the amplitude.

hep-lat↗

arXiv2023

Leading Power Accuracy in Lattice Calculations of Parton Distributions

In lattice-QCD calculations of parton distribution functions (PDFs) via large-momentum effective theory, the leading power (twist-three) correction appears as ${\cal O}(Λ_{\rm QCD}/P^z)$ due to the linear-divergent self-energy of Wilson line in quasi-PDF operators. For lattice data with hadron momentum $P^z$ of a few GeV, this correction is dominant in matching, as large as 30\% or more. We show how to eliminate this uncertainty through choosing the mass renormalization parameter consistently with the resummation scheme of the infrared-renormalon series in perturbative matching coefficients. An example on the lattice pion PDF data at $P^z = 1.9$ GeV shows an improvement of matching accuracy by a factor of more than $3\sim 5$ in the expansion region $x= 0.2\sim 0.5$.

hep-lat↗

arXiv2023

Threshold resummation for computing large-$x$ parton distribution through large-momentum effective theory

Parton distribution functions (PDFs) at large $x$ are poorly constrained by high-energy experimental data, but extremely important for probing physics beyond standard model at colliders. We study the calculation of PDFs at large-$x$ through large-momentum $P^z$ expansion of the lattice quasi PDFs. Similar to deep-inelastic scattering, there are two distinct perturbative scales in the threshold limit where the matching coefficient can be factorized into a space-like jet function at scale $P^z|1-y|$ and a pair of heavy-light Sudakov form factors at scale $P^z$. The matching formula allows us to derive a full renormalization group resummation of large threshold logarithms, and the result is consistent with the known calculation to the next-to-next to leading order (NNLO). This paves the way for direct large-$x$ PDFs calculations in lattice QCD. As by-products, we find that the space-like jet function is related to a time-like version calculated previously through analytic continuation, and the heavy-light Sudakov form factor, calculated here to NNLO, is a universal object appearing as well in the large momentum expansion of quasi transverse-momentum-dependent PDFs and quasi wave-function amplitudes.

hep-ph↗

arXiv2023

Resumming Quark's Longitudinal Momentum Logarithms in LaMET Expansion of Lattice PDFs

In the large-momentum expansion for parton distribution functions (PDFs), the natural physics scale is the longitudinal momentum ($p_z$) of the quarks (or gluons) in a large-momentum hadron. We show how to expose this scale dependence through resumming logarithms of the type $\ln^n p_z/μ$ in the matching coefficient, where $μ$ is a fixed renormalization scale. The result enhances the accuracy of the expansion at moderate $p_z>1$ GeV, and at the same time, clearly shows that the partons cannot be approximated from quarks with $p_z\sim Λ_{\rm QCD}$ which are not predominantly collinear with the parent hadron momentum, consistent with power counting of the large-momentum effective theory. The same physics mechanism constrains the coordinate space expansion at large distances $z$, the conjugate of $p_z$, as illustrated in the example of fitting the moments of the PDFs.

hep-ph↗

arXiv2023

Nucleon Transversity Distribution in the Continuum and Physical Mass Limit from Lattice QCD

We report a state-of-the-art lattice QCD calculation of the isovector quark transversity distribution of the proton in the continuum and physical mass limit using large-momentum effective theory. The calculation is done at four lattice spacings $a=\{0.098,0.085,0.064,0.049\}$~fm and various pion masses ranging between $220$ and $350$ MeV, with proton momenta up to $2.8$ GeV. The result is non-perturbatively renormalized in the hybrid scheme with self renormalization which treats the infrared physics at large correlation distance properly, and extrapolated to the continuum, physical mass and infinite momentum limit. We also compare with recent global analyses for the nucleon isovector quark transversity distribution.

hep-lat↗

arXiv2023

Transverse-Momentum-Dependent Wave Functions of Pion from Lattice QCD

We present a first lattice QCD calculation of the transverse-momentum-dependent wave functions (TMDWFs) of the pion using large-momentum effective theory. Numerical simulations are based on one ensemble with 2+1+1 flavors of highly improved staggered quarks action with lattice spacing $a=0.121$~fm from the MILC Collaboration, and one with 2 +1 flavor clover fermions and tree-level Symanzik gauge action generated by the CLS Collaboration with $a=0.098$~fm. As a key ingredient, the soft function is first obtained by incorporating the one-loop perturbative contributions and a proper normalization. Based on this and the equal-time quasi-TMDWFs simulated on the lattice, we extract the light-cone TMDWFs. The results are comparable between the two lattice ensembles and a comparison with phenomenological parametrization is made. Our studies provide a first attempt of $ab$ $initio$ calculation of TMDWFs which will eventually lead to crucial theory inputs for making predictions for exclusive processes under QCD factorization.

hep-lat↗

arXiv2022

Nonperturbative Determination of Collins-Soper Kernel from Quasi Transverse-Momentum Dependent Wave Functions

In the framework of large-momentum effective theory at one-loop matching accuracy, we perform a lattice calculation of the Collins-Soper kernel which governs the rapidity evolution of transverse-momentum-dependent (TMD) distributions. We first obtain the quasi TMD wave functions at three different meson momenta on a lattice with valence clover quarks on a dynamical HISQ sea and lattice spacing $a=0.12$~fm from MILC, and renormalize the pertinent linear divergences using Wilson loops. Through one-loop matching to the light-cone wave functions, we determine the Collins-Soper kernel with transverse separation up to 0.6~fm. We study the systematic uncertainties from operator mixing and scale dependence, as well as the impact from higher power corrections. Our results potentially allow for a determination of the soft function and other transverse-momentum dependent quantities at one-loop accuracy.

hep-lat↗

arXiv2022

Pion and Kaon Distribution Amplitudes from Lattice QCD

We present the state-of-the-art lattice QCD calculation of the pion and kaon light-cone distribution amplitudes (DAs) using large-momentum effective theory. The calculation is done at three lattice spacings $a\approx\{0.06,0.09,0.12\}$ fm and physical pion and kaon masses, with the meson momenta $P_z = \{1.29,1.72,2.15\}$ GeV. The result is non-perturbatively renormalized in a recently proposed hybrid scheme with self renormalization, and extrapolated to the continuum as well as the infinite momentum limit. We find a significant deviation of the pion and kaon DAs from the asymptotic form, and a large $SU(3)$ flavor breaking effect in the kaon DA.

hep-lat↗

arXiv2021

Self-Renormalization of Quasi-Light-Front Correlators on the Lattice

In applying large-momentum effective theory, renormalization of the Euclidean correlators in lattice regularization is a challenge due to linear divergences in the self-energy of Wilson lines. Based on lattice QCD matrix elements of the quasi-PDF operator at lattice spacing $a$= 0.03 fm $\sim$ 0.12 fm with clover and overlap valence quarks on staggered and domain-wall sea, we design a strategy to disentangle the divergent renormalization factors from finite physics matrix elements, which can be matched to a continuum scheme at short distance such as dimensional regularization and minimal subtraction. Our results indicate that the renormalization factors are universal in the hadron state matrix elements. Moreover, the physical matrix elements appear independent of the valence fermion formulations. These conclusions remain valid even with HYP smearing which reduces the statistical errors albeit reducing control of the renormalization procedure. Moreover, we find a large non-perturbative effect in the popular RI/MOM and ratio renormalization scheme, suggesting favor of the hybrid renormalization procedure proposed recently.

hep-lat↗

arXiv2020

Privacy-preserving Learning via Deep Net Pruning

This paper attempts to answer the question whether neural network pruning can be used as a tool to achieve differential privacy without losing much data utility. As a first step towards understanding the relationship between neural network pruning and differential privacy, this paper proves that pruning a given layer of the neural network is equivalent to adding a certain amount of differentially private noise to its hidden-layer activations. The paper also presents experimental results to show the practical implications of the theoretical finding and the key parameter values in a simple practical setting. These results show that neural network pruning can be a more effective alternative to adding differentially private noise for neural networks.

cs.LG↗

arXiv2019

Two-mediator dark matter models and cosmic electron excess

The cosmic electron energy spectrum recently observed by the DAMPE experiment exhibits two interesting features, including a break around 0.9 TeV and a sharp resonance near 1.4 TeV. In this analysis, we propose a dark matter explanation to both exotic features seen by DAMPE. In our model, dark matter annihilates in the galaxy via two different channels that lead to both a narrow resonance spectrum near 1.4 TeV and electron excess events over an extended energy range thus generating the break structure around TeV. The two annihilation channels are mediated by two gauge bosons that interact both with dark matter and with the standard model fermions. Dark matter annihilations through the s-channel process mediated by the heavier boson produce monoenergetic electron-positron pairs leading to the resonance excess. The lighter boson has a mass smaller than the dark matter such that they can be on-shell produced in dark matter annihilations in the galaxy; the lighter bosons in the final state subsequently decay to generate the extended excess events due to the smeared electron energy spectrum in this process. We further analyze constraints from various experiments, including HESS, Fermi, AMS, and LHC, to the parameter space of the model where both excess events can be accounted for. In order to interpret the two new features in the DAMPE data, dark matter annihilation cross sections in the current galaxy are typically much larger than the canonical thermal cross section needed for the correct dark matter relic abundance. This discrepancy, however, is remedied by the nonperturbative Sommerfeld enhancement because of the existence of a lighter mediator in the model.

hep-ph↗

arXiv2017

A combined analysis of PandaX, LUX, and XENON1T experiments within the framework of dark matter effective theory

Weakly interacting massive particles are a widely well-probed dark matter candidate by the dark matter direct detection experiments. Theoretically, there are a large number of ultraviolet completed models that consist of a weakly interacting massive particle dark matter. The variety of models makes the comparison with the direct detection data complicated and often non-trivial. To overcome this, in the non-relativistic limit, the effective theory was developed in the literature which works very well to significantly reduce the complexity of dark matter-nucleon interactions and to better study the nuclear response functions. In the effective theory framework for a spin-1/2 dark matter, we combine three independent likelihood functions from the latest PandaX, LUX, and XENON1T data, and give a joint limit on each effective coupling. The astrophysical uncertainties of the dark matter distribution are also included in the likelihood. We further discuss the isospin violating cases of the interactions. Finally, for both dimension-five and dimension-six effective theories above the electroweak scale, we give updated limits of the new physics mass scales.

hep-ph↗