arXiv subjects
Publications and source records attributed to Urs Wenger.
We present a simulation algorithm for Wilson fermions based on the exact hopping expansion of the fermion action. The algorithm essentially eliminates critical slowing down by sampling the fermionic two-point correlation function and it allows simulations directly in the massless limit. As illustrative examples, the algorithm is applied to the Gross-Neveu and the Schwinger model, the latter in the strong coupling limit.
We have developed an efficient simulation algorithm for strongly interacting relativistic fermions in two-dimensional field theories based on a formulation as a loop gas. The loop models describing the dynamics of the fermions can be mapped to statistical vertex models and our proposal is in fact an efficient simulation algorithm for generic vertex models in arbitrary dimensions. The algorithm essentially eliminates critical slowing down by sampling two-point correlation functions and it allows simulations directly in the massless limit. Moreover, it generates loop configurations with fluctuating topological boundary conditions enabling to simulate fermions with arbitrary periodic or anti-periodic boundary conditions. As illustrative examples, the algorithm is applied to the Gross-Neveu model and to the Schwinger model in the strong coupling limit.
We present the status of runs performed in the twisted mass formalism with $N_{\rm f}=2+1+1$ flavours of dynamical fermions: a degenerate light doublet and a mass split heavy doublet. The procedure for tuning to maximal twist will be described as well as the current status of the runs using both thin and stout links. Preliminary results for a few observables obtained on ensembles at maximal twist will be given. Finally, a reweighting procedure to tune to maximal twist will be described.
We investigate the Gross-Neveu model on the lattice at finite temperature and chemical potential in the limit of an infinite number of fermion flavours. We check the universality of the continuum limit of staggered and overlap fermions at finite temperature and chemical potential. We show that at finite density a recently discovered phase of cold baryonic matter emerges as a baryon crystal from a spatially inhomogeneous fermion condensate. However, we also demonstrate that on the lattice, this new phase disappears at large coupling or in small volumes. Furthermore, we investigate unusual finite size effects that appear at finite chemical potential. Finally, we speculate on the implications of our findings for QCD.
Some key features of the overlap operator with a UV-filtered Wilson kernel are discussed. The first part concerns spectral properties of the underlying shifted hermitean Wilson operator and the relation to the observed speedup of the overlap construction. Next, the localization of the filtered overlap and its axial-vector renormalization constant are discussed. Finally, results of an exploratory scaling study for $m_{ud}, m_s$ and $f_π, f_K$ are presented.
We present a comparison of various five-dimensional representations of chiral fermions considering their cost and residual chiral symmetry breaking.
We discuss the kernel spectrum, locality properties and the axial-vector renormalization constant of UV-filtered overlap fermions. We find that UV-filtered overlap fermions have a better conditioned kernel, better locality and an axial-vector renormalization constant closer to 1 than their unfiltered counterparts, even if the shift parameter $ρ$ is simply set to 1.
We investigate the overlap operator with a UV filtered Wilson kernel. The filtering leads to a better localization of the operator even on coarse lattices and with the untuned choice $ρ=1$. Furthermore, the axial-vector renormalization constant $Z_A$ is much closer to 1, reducing the mismatch with perturbation theory. We show that all these features persist over a wide range of couplings and that the details of filtering prove immaterial. We investigate the properties of the kernel spectrum and find that the kernel non-normality is reduced. As a side effect we observe that for certain applications of the filtered overlap a speed-up factor of 2-4 can be achieved.
We study the infrared part of the spectrum for UV-filtered staggered Dirac operators and compare them to the overlap counterpart. With sufficient filtering and at small enough lattice spacing the staggered spectra manage to ``mimic'' the overlap version. They show a 4-fold near-degeneracy, and a clear separation between would-be zero modes and non-zero modes. This suggests an approximate index theorem for filtered staggered fermions and a correct sensitivity to the topology of QCD. Moreover, it supports square-rooting the staggered determinant to obtain dynamical ensembles with $N_f=2$.
We test a variety of blocking and smearing algorithms for constructing glueball and string wave-functionals, and find some with much improved overlaps onto the lightest states. We use these algorithms to obtain improved results on the tensions of k-strings in SU(4), SU(6), and SU(8) gauge theories. We emphasise the major systematic errors that still need to be controlled in calculations of heavier k-strings, and perform calculations in SU(4) on an anisotropic lattice in a bid to minimise one of these. All these results point to the k-string tensions lying part-way between the `MQCD' and `Casimir Scaling' conjectures, with the power in 1/N of the leading correction lying in [1,2]. We also obtain some evidence for the presence of quasi-stable strings in calculations that do not use sources, and observe some near-degeneracies between (excited) strings in different representations. We also calculate the lightest glueball masses for N=2, ...,8, and extrapolate to N=infinity, obtaining results compatible with earlier work. We show that the N=infinity factorisation of the Euclidean correlators that are used in such mass calculations does not make the masses any less calculable at large N.
We use a continued fraction expansion of the sign-function in order to obtain a five dimensional formulation of the overlap lattice Dirac operator. Within this formulation the inverse of the overlap operator can be calculated by a single Krylov space method and nested conjugate gradient procedures are avoided. We point out that the five dimensional linear system can be made well conditioned using equivalence transformations on the continued fractions.
We calculate the topological charge density of SU(N) lattice gauge fields for values of N up to N=8. Our T=0 topological susceptibility appears to approach a finite non-zero limit at N=infinity that is consistent with earlier extrapolations from smaller values of N. Near the deconfining temperature Tc we are able to investigate separately the confined and deconfined phases, since the transition is quite strongly first order. We find that the topological susceptibility of the confined phase is always very similar to that at T=0. By contrast, in the deconfined phase at larger N there are no topological fluctuations except for rare, isolated and small instantons. This shows that as N->infinity the large-T suppression of large instantons and the large-N suppression of small instantons overlap, even at T=Tc, so as to suppress all topological fluctuations in the deconfined phase. In the confined phase by contrast, the size distribution is much the same at all T, becoming more peaked as N grows, suggesting that D(rho) is proportional to a delta function at N=infinity, centered on rho close to 1/Tc.
We calculate the continuum value of the deconfining temperature in units of the string tension for SU(4), SU(6) and SU(8) gauge theories, and we recalculate its value for SU(2) and SU(3). We find that the $N$-dependence for $2 \leq N \leq 8$ is well fitted by $T_c/\sqrt{sigma} = 0.596(4) + 0.453(30)/N^2$, showing a rapid convergence to the large-N limit. We confirm our earlier result that the phase transition is first order for $N \geq 3$ and that it becomes stronger with increasing $N$. We also confirm that as $N$ increases the finite volume corrections become rapidly smaller and the phase transition becomes visible on ever smaller volumes. We interpret the latter as being due to the fact that the tension of the domain wall that separates the confining and deconfining phases increases rapidly with $N$. We speculate on the connection to Eguchi-Kawai reduction and to the idea of a Master Field.
We study the low-lying eigenmodes of the lattice overlap Dirac operator for SU(N) gauge theories with N=2,3,4 and 5 colours. We define a fermionic topological charge from the zero-modes of this operator and show that, as N grows, any disagreement with the topological charge obtained by cooling the fields, becomes rapidly less likely. By examining the fields where there is a disagreement, we are able to show that the Dirac operator does not resolve instantons below a critical size of about rho = 2.5 a, but resolves the larger, more physical instantons. We investigate the local chirality of the near-zero modes and how it changes as we go to larger N. We observe that the local chirality of these modes, which is prominent for SU(2) and SU(3), becomes rapidly weaker for larger N and is consistent with disappearing entirely in the limit of N -> infinity. We find that this is not due to the observed disappearance of small instantons at larger N.
We present a method for constructing classically perfect anisotropic actions for SU(3) gauge theory based on an isotropic Fixed Point Action. The action is parametrised using smeared (``fat'') links. The construction is done explicitly for anisotropy $ξ=a_s/a_t=2$ and 4. The corresponding renormalised anisotropies are determined using the torelon dispersion relation. The renormalisation of the anisotropy is small and the parametrisation describes the true action well. Quantities such as the static quark-antiquark potential, the critical temperature of the deconfining phase transition and the low-lying glueball spectrum are measured on lattices with anisotropy $ξ=2$. The mass of the scalar $0^{++}$ glueball is determined to be 1580(60) MeV, while the tensor $2^{++}$ glueball is at 2430(60) MeV.
We present a new parametrization of a SU(3) fixed point (FP) gauge action using smeared ("fat") gauge links. We report on the scaling behaviour of the FP action on coarse lattices by means of the static quark-antiquark potential, the hadronic scale $r_0$, the string tension $σ$ and the critical temperature $T_c$ of the deconfining phase transition. In addition, we investigate the low lying glueball masses where we observe no scaling violations within the statistical errors.
On anisotropic lattices, where generally the lattice is rather coarse in spatial directions, a parametrized classically perfect action could help reducing lattice artifacts considerably. We investigate the possibility of constructing such actions for SU(3) gauge theory. We present two different methods to do so, either repeating the procedure used to create our newly parametrized isotropic FP action, or performing one single step starting with the isotropic result. The anisotropic action is parametrized using an ansatz including anisotropically APE-like smeared (``fat'') links. The parametrized classically perfect action with anisotropy $ξ=a_s/a_t=2$ is constructed and the renormalized anisotropy is measured using the torelon dispersion relation. It turns out that the renormalization is small.
A new parametrization is introduced for the fixed point (FP) action in SU(3) gauge theory using fat links. We investigate its scaling properties by means of the static quark-antiquark potential and the dimensionless quantities $r_0 T_c, T_c/\sqrtσ$ and $r_0 \sqrtσ$, where $T_c$ is the critical temperature of the deconfining phase transition, $r_0$ is the hadronic scale and $σ$ is the effective string tension. These quantities scale even on lattices as coarse as $a \approx 0.3$ fm. We also measure the glueball spectrum and obtain $m_{0^{++}}=1627(83)$ MeV and $m_{2^{++}}=2354(95)$ MeV for the masses of the scalar and tensor glueballs, respectively.