Search

We show the equivalence of the 2D Ising model to standard free Euclidean lattice fermions of the Wilson Majorana type. The equality of the loop representations for the partition functions of both systems is established exactly for finite lattices with well-defined boundary conditions. The honeycomb lattice is particularly simple in this context and therefore discussed first and only then followed by the more familiar square lattice case., Comment: 19 pages, 5 figures, sec. 3.3 added detailing the relation with Samuel's fermionization

We discuss the problem of lattice artefacts in QCD simulations enhanced by the introduction of dynamical charmed quarks. In particular, we advocate the use of a massive renormalization scheme with a close to realistic charm mass. To maintain O(a) improvement for Wilson type fermions in this case we define a finite size scheme and carry out a nonperturbative estimation of the clover coefficient $c_\mathrm{sw}$. It is summarized in a fit formula $c_\mathrm{sw}(g_0^2)$ that defines an improved action suitable for future dynamical charm simulations., Comment: 17 pages, 3 tables, 2 figures, published version (extended introduction)

We study the O(3) sigma model in $D=2$ on the lattice with a Boltzmann weight linearized in $\beta$ on each link. While the spin formulation now suffers from a sign-problem the equivalent loop model remains positive and becomes particularly simple. By studying the transfer matrix and by performing Monte Carlo simulations in the loop form we study the mass gap coupling in a step scaling analysis. The question addressed is, whether or not such a simplified action still has the right universal continuum limit. If the answer is affirmative this would be helpful in widening the applicability of worm algorithm methods., Comment: 7 pages, 7 figures, Lattice 2016 (Southampton, UK)

We define a finite size renormalization scheme for $\phi^4$ theory which in the thermodynamic limit reduces to the standard scheme used in the broken phase. We use it to re-investigate the question of triviality for the four dimensional infinite bare coupling (Ising) limit. The relevant observables all rely on two-point functions and are very suitable for a precise estimation with the worm algorithm. This contribution updates an earlier publication by analysing a much larger dataset., Comment: 6 pages, 2 figures, 1 table, presented at the 32nd International Symposium on Lattice Field Theory (Lattice 2014), June 23-28 2014, New York, USA

We develop a strategy for the non-perturbative determination of the O(a)-improvement coefficient c_sw for Wilson fermions with massive sea quarks. The improvement condition is defined via the PCAC relation in the Schr\"odinger functional. It is imposed along a line of constant physics designed to be close to the correct mass of the charm quark. The numerical work uses the tree-level improved L\"uscher-Weisz gauge action in N_f=3+1 Lattice QCD.

We review the long term project of the ALPHA collaboration to compute in QCD the running coupling constant and quark masses at high energy scales in terms of low energy hadronic quantities. The adapted techniques required to numerically carry out the required multiscale non-perturbative calculation with our special emphasis on the control of systematic errors are summarized. The complete results in the two dynamical flavor approximation are reviewed and an outlook is given on the ongoing three flavor extension of the programme with improved target precision., Comment: 30 pages, 20 figures, Contribution to the Proceedings of the Final Meeting of DFG SFB-TR-9 (Durbach, Germany, Sept. 2014), to appear in Nucl. Phys. (Proc. Suppl.)

We study the standard one-component $\varphi^4$-theory in four dimensions. A renormalized coupling is defined in a finite size renormalization scheme which becomes the standard scheme of the broken phase for large volumes. Numerical simulations are reported using the worm algorithm in the limit of infinite bare coupling. The cutoff dependence of the renormalized coupling closely follows the perturbative Callan Symanzik equation and the triviality scenario is hence further supported., Comment: 10 pages, 2 figures

We present a Monte-Carlo algorithm for the simulation of the all-order strong coupling expansion of the Z2 gauge theory. This random surface ensemble is equivalent to the standard formulation, but allows to measure some quantities, like Polyakov loop correlators or excess free energies, with an accuracy that could not have been easily achieved with traditional simulation methods. One interesting application of the algorithm is the investigation of effective string theories., Comment: presented at the 31st International Symposium on Lattice Field Theory (Lattice 2013), 29 July - 3 August 2013, Mainz, Germany

All-order strong coupling simulations have been used to derive precise energies of string states in the confined phase of three dimensional Z(2) lattice gauge theory. The behavior of the ground state energy is here compared with predictions of effective string theory. Our new data reported here are consistent with known universal terms of the long string length ($L_0$) expansion known from effective string models in the continuum limit. Our precision is however still not sufficient to disentangle non-univeral terms of order $L_0^{-7}, Comment: 7 pages, 1 figure, 2 tables. PoS (Lattice 2013) 2013

We exactly rewrite the Z(2) lattice gauge theory with standard plaquette action as a random surface model equivalent to the untruncated set of its strong coupling graphs. By extending the worm approach applied to spin models we simulate such surfaces including Polyakov line defects that randomly walk over the lattice. Our Monte Carlo algorithms for the graph ensemble are reasonably efficient but not free of critical slowing down. Polyakov line correlators can be measured in this approach with small relative errors that are independent of the separation. As a first application our results are confronted with effective string theory predictions. In addition, the excess free energy due to twisted boundary conditions becomes an easily accessible observable. Our numerical experiments are in three dimensions, but the method is expected to work in any dimension., Comment: 25 pages, 7 figures, V2: small changes in the text, refs. added, new footnote p.18, to appear in Nucl. Phys. B [FS]

We represent low dimensional quantum mechanical Hamiltonians by moderately sized finite matrices that reproduce the lowest O(10) boundstate energies and wave functions to machine precision. The method extends also to Hamiltonians that are neither Hermitian nor PT symmetric and thus allows to investigate whether or not the spectra in such cases are still real. Furthermore, the approach is especially useful for problems in which a position-dependent mass is adopted, for example in effective-mass models in solid-state physics or in the approximate treatment of coupled nuclear motion in molecular physics or quantum chemistry. The performance of the algorithm is demonstrated by considering the inversion motion of different isotopes of ammonia molecules within a position-dependent-mass model and some other examples of one- and two-dimensional Hamiltonians that allow for the comparison to analytical or numerical results in the literature., Comment: 10 pages, 5 figures. Several clarifications in the text and new sect. IV.G. Version to appear in Phys. Rev. E

Worm methods to simulate the Ising model in the Aizenman random current representation including a low noise estimator for the connected four point function are extended to allow for antiperiodic boundary conditions. In this setup several finite size renormalization schemes are formulated and studied with regard to the triviality of \phi^4 theory in four dimensions. With antiperiodicity eliminating the zero momentum Fourier mode a closer agreement with perturbation theory is found compared to the periodic torus., Comment: 20 pages, 3 double-figures, 6 tables

We describe an explicit algorithm to factorize an even antisymmetric N^2 matrix into triangular and trivial factors. This allows for a straight forward computation of Pfaffians (including their signs) at the cost of N^3/3 flops., Comment: 6 pages, 1 figure, V2: Minor changes in the text and refs. added, to appear in CPC

Worm algorithms have been very successful with the simulation of sigma models with fixed length spins which result from scalar field theories in the limit of infinite quartic coupling lambda. Here we investigate closer their algorithmic efficiency at finite and even vanishing lambda for the one component model in dimensions D = 2, 3, 4., Comment: 10 pages, 2 Figs

We study a renormalized coupling g and mass m in four dimensional phi^4 theory on tori with finite size z=mL. Precise numerical values close to the continuum limit are reported for z=1,2,4, based on Monte Carlo simulations performed in the equivalent all-order strong coupling reformulation. Ordinary renormalized perturbation theory is found to work marginally at z=2 and and to fail at z=1. By exactly integrating over the constant field mode we set up a renormalized expansion in z and compute three nontrivial orders. These results reasonably agree with the numerical data at small z. In the new expansion, the universal continuum limit exists as expected from multiplicative renormalizability. The triviality scenario is corroborated with significant precision., Comment: 27 pages, 7 figures, 4 tables

We discuss the status of different determinations of alpha_s, motivating a precise and reliable computation from lattice QCD. In order to suppress perturbative errors, the non-perturbative computation has to reach high energy scales mu. Such results already exist in the SF-scheme for N_f=0,2 and N_f=3. We recently added the running with four massless flavours in a range of alpha from about 0.07 to 0.3 . It is based on our recent determination of the Sheikholeslami Wohlert coefficient in the four-flavour theory., Comment: 7 pages, Lattice 2010 proceedings

We present an algorithm in which the all-order strong coupling expansion of the Abelian U(1) gauge theory with Wilson plaquette action is sampled. In addition to the vacuum closed surface graphs of the partition function we propose to also allow for a class of defects (boundaries) related to Wilson loops in the ensemble. The efficiency of our scheme in estimating various observables is compared to a standard Metropolis algorithm., Comment: 7 pages, 2 figures. Presented at the XXVIII International Symposium on Lattice Field Theory (Lattice 2010), June 14-19 2010, Villasimius, Italy

We give an overview on recently accomplished successful generalizations of `worm' or loop gas simulation methods to O(N) and CP(N-1) sigma models and to simple fermion models. Beside the advantage of (practically) eliminated critical slowing down we also explain additional opportunities to estimate some observables with extremely improved signal to noise levels., Comment: Contribution to the proceedings of Lattice 2010, Sardinia, 14 pages, 7 figures

We have calculated the step scaling function and the running coupling of QCD in the Schroedinger functional scheme with four flavors of O(a) improved Wilson quarks. Comparisons of our non-perturbative results with 2-loop and 3-loop perturbation theory as well as with non-perturbative data for only two flavors are made., Comment: 22 pages, 4 figures

We exactly reformulate the lattice CP(N-1) spin model on a D dimensional torus as a loop model whose configurations correspond to the complete set of strong coupling graphs of the original system. A Monte Carlo algorithm is described and tested that samples the loop model with its configurations stored and manipulated as a linked list. Complete absence of critical slowing down and correspondingly small errors are found at D=2 for several observables including the mass gap. Using two different standard lattice actions universality is demonstrated in a finite size scaling study. The topological charge is identified in the loop model but not yet investigated numerically., Comment: 24 pages, 4 figures. The combinatoric derivation of the (unchanged) algorithm is corrected.

We have determined the non-perturbative O(a)-improvement coefficient c_sw for four flavors of Wilson quarks with the plaquette gauge action in a range of beta >= 5. The data are fitted with several Pade approximation formulae to get an impression of the stability. A small extrapolation below beta=5 seems acceptable., Comment: 15 pages, 4 figures

We reformulate the O(N) sigma model as a loop model whose configurations are the all-order strong coupling graphs of the original model. The loop configurations are represented by a pointer list in the computer and a Monte Carlo update scheme is proposed. Sample simulations are reported and the method turns out to be similarly efficient as the reflection cluster method, but it has greater potential for systematic generalization to other lattice field theories. A variant action suggested by the method is also simulated and leads to a rather extreme demonstration of the concept of universality of the scaling or continuum limit., Comment: 27 pages, 6 figures, 5 tables. The combinatoric derivation of the (unchanged) algorithm is corrected. Independently, all data in sect.5 are revised due to a software error, with unchanged conclusions on universality.

We investigate the discrete chiral transformation of a Majorana fermion on a torus. Depending on the boundary conditions the integration measure can change sign. Taking this anomalous behavior into account we define a chiral order parameter as a ratio of partition functions with differing boundary conditions. Then the lattice realization of the Gross-Neveu model with Wilson fermions is simulated using the recent `worm' technique on the loop gas or all-order hopping representation of the fermions. An algorithm is formulated that includes the Gross-Neveu interaction for N fermion species. The critical line m_c(g) is constructed for a range of couplings at N = 6 and for N = 2, the Thirring model, as examples., Comment: 21 pages, 1 figure. Typos corrected and a few sentences improved. To appear in Nucl. Phys. B [FS]

We report precise simulations of phi^4 theory in the Ising limit. A recent technique to stochastically evaluate the all-order strong coupling expansion is combined with exact identities in the closely related Aizenman random current representation. In this way high precision estimates of the renormalized coupling are possible at low CPU cost. As a sample application we present results for the unbroken phase of the Ising model in dimensions 3, 4 and 5 and investigate the question of triviality by studying a finite size scaling continuum limit., Comment: 10 pages, 3 figures, Introductory paragraph added, one line missing in the table added, to appear in Phys. Rev. D

We investigate the extension of the Prokof'ev-Svistunov worm algorithm to Wilson lattice fermions in an external scalar field. We effectively simulate by Monte Carlo the graphs contributing to the hopping expansion of the two-point function on a finite lattice to arbitrary order. Tests are conducted for a constant background field i. e. free fermions at some mass. For the method introduced here this is expected to be a representative case. Its advantage is that we know the exact answers and can thus make stringent tests on the numerics. The approach is formulated in both two and three space-time dimensions. In D=2 Wilson fermions enjoy special positivity properties and the simulation is similarly efficient as in the Ising model. In D=3 the method also works at sufficiently large mass, but there is a hard sign problem in the present formulation hindering us to take the continuum limit., Comment: 29 pages [12pt], 5 figures

We investigate in some detail an alternative simulation strategy for lattice field theory based on the so-called worm algorithm introduced by Prokof'ev and Svistunov in 2001. It amounts to stochastically simulating the strong coupling expansion rather than the usual configuration sum. A detailed error analysis and an important generalization of the method are exemplified here in the simple Ising model. It allows for estimates of the two point function where in spite of exponential decay the signal to noise ratio does not degrade at large separation. Critical slowing down is practically absent. In the outlook some thoughts on the general applicability of the method are offered., Comment: 15 pages, 2 figures, refs. added, small language changes, to app. in Nucl. Phys. B[FS]

We report on a scaling test of several mesonic observables in the non-perturbatively O(a) improved Wilson theory with two flavors of dynamical quarks. The observables are constructed in a fixed volume of 2.4fm x (1.8fm)^3 with Schroedinger functional boundary conditions. No significant scaling violations are found. Using the kaon mass determined in \cite{cernI}, we update our estimate of the Lambda parameter to Lambda^(2)_{msbar}/m_K = 0.52(6)., Comment: 12 pages, 5 tables, 3 figures; one equation corrected; last line and caption of table 4 revised

The (discrete) Gross-Neveu model is studied in a lattice realization with an N-component Majorana Wilson fermion field. It has an internal O(N) symmetry in addition to the euclidean lattice symmetries. The discrete chiral symmetry for vanishing mass is expected to emerge in the continuum limit only. The lattice theory is first recast in terms of two-valued bosonic link variables (dimers). In this representation, which coincides with the loop representation obtained earlier by Gattringer with the help of eight-vertex-models, the Boltzmann weight is essentially positive. While standard local updates are possible in this form we construct a further exact transformation where we generate dimer configurations as Peierls contours of an Ising model with a local action residing on plaquettes. For this model a Swendsen-Wang type cluster algorithm is constructed. At vanishing coupling it is numerically demonstrated to almost completely eliminate critical slowing down. Although further tests are required, an avenue to numerical studies of the Gross-Neveu model with unprecedented precision seems open., Comment: 7 pages, 2 figures, contribution to Lattice 2007 (Regensburg, Germany)

We report on some preparatory investigations for the simulation of the QCD Schroedinger functional with a non-hermitian polynomial hybrid Monte Carlo algorithm. The complex spectrum of the non-hermitean free operator with SF boundary condititons is computed semianalytically. Then it is shown how one can obtain relevant information on the boundary of the spectral domain also in the presence of nontrivial gaugefields by monitoring the behavior of polynomials in the Wilson operator applied on random vectors., Comment: 7 pages, 5 figures, 2 tables, poster presented at Lattice 2007, Regensburg, Germany

We present a multiplication algorithm to recursively construct vertices for the Schroedinger functional in the abelian background field case. The algorithm is suited for automatic perturbative calculations with a variety of actions. As first applications, we derive ratios of the lambda parameters between the lattice scheme (improved gauge actions including six link loops) and the $\bar{\rm MS}$ scheme, and one-loop results for the Schroedinger functional coupling with a lattice $T=L \pm a$, which is motivated by considering staggered fermions., Comment: 7 pages, 2 figures, 1 table, contribution to Lattice 2007, Regensburg, Germany

For Majorana-Wilson lattice fermions in two dimensions we derive a dimer representation. This is equivalent to Gattringer's loop representation, but is made exact here on the torus. A subsequent dual mapping leads to yet another representation in which a highly efficient Swendsen-Wang type cluster algorithm is constructed. It includes the possibility of fluctuating boundary conditions. It also allows for improved estimators and makes interesting new observables accessible to Monte Carlo. The algorithm is compatible with the Gross-Neveu as well as an additional Z(2) gauge interaction. In this article numerical demonstrations are reported for critical free fermions., Comment: 24 pages, 3 figures: tiny changes, mainly typos

We introduce a finite volume renormalization scheme for the N-Majorana-component O(N) invariant Gross-Neveu model. Universal observables are defined that are accessible to precise numerical simulation in various discretizations and allow for an extrapolation to the continuum limit. Here first numerical results with Wilson fermions are reported. For N=2 they reproduce exact finite volume continuum results in the massless Thirring model. Our N=8 data are ready for comparison for instance with staggered results in the future., Comment: 7 pages, 4 figures, contribution to Lattice 2006 (Tuczon/AZ/USA)

We study autocorrelation times of physical observables in lattice QCD as a function of the molecular dynamics trajectory length in the hybrid Monte-Carlo algorithm. In an interval of trajectory lengths where energy and reversibility violations can be kept under control, we find a variation of the integrated autocorrelation times by a factor of about two in the quantities of interest. Trajectories longer than conventionally used are found to be superior both in the Nf=0 and Nf=2 examples considered here. We also provide evidence that they lead to faster thermalization of systems with light quarks., Comment: 13 pages, 6 figures

We investigate the two flavor chiral Gross-Neveu model in the Schroedinger functional on the lattice. The procedure necessary to recover chiral symmetry in the continuum limit of this model with Wilson fermions is discussed. We introduce several useful observables and present a first demonstration of the feasibility of Monte-Carlo simulations in this model., Comment: Lattice 2005

The non-perturbative running of the quark mass in the Schroedinger functional scheme is computed over a large energy range (covering scales differing by two orders of magnitude). This allows to relate lattice estimates of the running quark mass to the renormalization group invariant mass. The result is used in a preliminary computation of the strange quark mass in the theory with two flavors of non-perturbatively improved Wilson quarks. A more detailed discussion of the calculation can be found in [1]., Comment: Talk presented at Lattice 2005 (Improvement and renomalisation), 6 pages, 2 figures

We present results for the non-perturbative determination of the improvement and renormalization factors of the isovector axial current for lattice QCD with two flavors of dynamical Wilson quarks. The improvement and normalization conditions are formulated in terms of matrix elements of the PCAC relation in the Schroedinger functional setup and results are given in the form of interpolating formulae for bare gauge couplings beta=6/g_0^2>5.2., Comment: Talk presented at Lattice 2005 (Improvement and renomalisation), 6 pages, 4 figures

The computation of the step scaling function for the finite size mass-gap in the O(N) sigma model at large N is reviewed. Practically exact nonperturbative results become available for both finite and vanishing lattice spacing. We use them as a testbed to investigate standard procedures of continuum extrapolation in lattice field theory., Comment: 6 pages, 2 figures, contribution of U.W. to Lat05, Dublin

The running of renormalized quark masses is computed in lattice QCD with two flavors of massless O(a) improved Wilson quarks. The regularization and flavor independent factor that relates running quark masses to the renormalization group invariant ones is evaluated in the Schroedinger Functional scheme. Using existing data for the scale r_0 and the pseudoscalar meson masses, we define a reference quark mass in QCD with two degenerate quark flavors. We then compute the renormalization group invariant reference quark mass at three different lattice spacings. Our estimate for the continuum value is converted to the strange quark mass with the help of chiral perturbation theory., Comment: 25 pages, 6 figures; sections 1 and 4 rearranged, minor change to the summary plot

In the nonlinear O(N) sigma model at N=3 unexpected cutoff effects have been found before with standard discretizations and lattice spacings. Here the situation is analyzed further employing additional data for the step scaling function of the finite volume massgap at N=3,4,8 and a large N-study of the leading as well as next-to-leading terms in 1/N. The latter exact results are demonstrated to follow Symanzik's form of the asymptotic cutoff dependence. At the same time, when fuzzed with artificial statistical errors and then fitted like the Monte Carlo results, a picture similar to N=3 emerges. We hence cannot conclude a truly anomalous cutoff dependence but only relatively large cutoff effects, where the logarithmic component is important. Their size shrinks at larger N, but the structure remains similar. The large N results are particularly interesting as we here have exact nonperturbative control over an asymptotically free model both in the continuum limit and on the lattice., Comment: 23 pages, 13 Figures, minor changes for clarification

We present a new normalization condition for the axial current, derived from the PCAC relation with non-vanishing quark mass. This condition is expected to reduce mass effects in the chiral extrapolation of the results for the normalization factor Z_A. The application to the two-flavor theory with improved Wilson fermions shows that this expectation is indeed fulfilled. Using the Schroedinger functional setup we calculate Z_A(g_0^2) as well as the vector current normalization factor Z_V(g_0^2) for beta = 6/g_0^2 >= 5.2., Comment: 15 pages, 4 figures, 2 tables, JHEP style

We present a non-perturbative computation of the running of the coupling alpha_s in QCD with two flavours of dynamical fermions in the Schroedinger functional scheme. We improve our previous results by a reliable continuum extrapolation. The Lambda-parameter characterizing the high-energy running is related to the value of the coupling at low energy in the continuum limit. An estimate of Lambda*r_0 is given using large-volume data with lattice spacings a from 0.07 fm to 0.1 fm. It translates into Lambda_{MSbar}^{(2)}=245(16)(16) MeV [assuming r_0=0.5 fm]. The last step still has to be improved to reduce the uncertainty., Comment: 34 pages including figures and tables, latex2e

As a feasibility study for a scaling test we investigate the behavior of algorithms for dynamical fermions in the N_f=2 Schroedinger functional at an intermediate volume of 1 fm^4. Simulations were performed using HMC with two pseudo-fermions and PHMC at lattice spacings of approximately 0.1 and 0.07 fm. We show that some algorithmic problems are due to large cutoff-effects in the spectrum of the improved Wilson-Dirac operator and disappear at the smaller lattice spacing. The problems discussed here are not expected to be specific to the Schroedinger functional., Comment: 19 pages, 12 figures, Sec. 2 extended, few references added. Accepted for publication in Comp. Phys. Comm

Recent exact predictions for the massive scaling limit of the two dimensional XY-model are based on the equivalence with the sine-Gordon theory and include detailed results on the finite size behavior. The so-called step-scaling function of the mass gap is simulated with very high precision and found consistent with analytic results in the continuum limit. To come to this conclusion, an also predicted form of a logarithmic decay of lattice artifacts was essential to use for the extrapolation., Comment: 16 pages, 3 figures

A study of principle is conducted on the inclusion of the fermionic determinant as a Metropolis acceptance correction. It is carried out in the 2-D Schwinger model to prepare later applications to the Schr"odinger functional. A mixed stochastic/determistic acceptance step is found that allows to include some problematic modes in a way to avoid the collapse of the acceptance rate due to fluctuations., Comment: Lattice2003(machine), 3 pages, 3 Figures, talk by U.W

We explain in detail how to estimate mean values and assess statistical errors for arbitrary functions of elementary observables in Monte Carlo simulations. The method is to estimate and sum the relevant autocorrelation functions, which is argued to produce more certain error estimates than binning techniques and hence to help toward a better exploitation of expensive simulations. An effective integrated autocorrelation time is computed which is suitable to benchmark efficiencies of simulation algorithms with regard to specific observables of interest. A Matlab code is offered for download that implements the method. It can also combine independent runs (replica) allowing to judge their consistency., Comment: 22 pages, 4 figures, link-address for software download, V4: Improvement in eq.(58) and (42) for error of tau_int => new version of software.Only subleading error terms affected, results should remain compatible

In the simplified setting of the Schwinger model we present a systematic study on the simulation of dynamical fermions by global accept/reject steps that take into account the fermion determinant. A family of exact algorithms is developed, which combine stochastic estimates of the determinant ratio with the exploitation of some exact extremal eigenvalues of the generalized problem defined by the `old' and the `new' Dirac operator. In this way an acceptable acceptance rate is achieved with large proposed steps and over a wide range of couplings and masses., Comment: 39 pages, 9 figures, small changes in the text, Fig.5 and Tab.2 (incl. 1 corrected typo)

We study the Schr\"odinger functional coupling for lattice Yang-Mills theory coupled to an improved bosonic spinor field, which corresponds to QCD with minus two light flavors. This theory serves as a less costly testcase than QCD for the scaling of the coupling., Comment: Lattice2001(improvement) 3 pages, 4 figures

High precision measurements of the renormalized zero-momentum 4-point coupling g_R and of the Luscher-Weisz-Wolff running coupling gbar(L) = L*m(L) performed with two different lattice actions in the non-perturbative region confirm the earlier observations, that the cutoff effects look linear, in contrast to perturbative considerations. The use of different actions allows one to make a more reliable estimate on the continuum limit. The measurements were done for infinite volume correlation length up to 350., Comment: 3 pages, 2 figures, Lattice2001(spin)

Existing non-perturbative computations of the running coupling of quenched QCD in the Schroedinger functional scheme are extended to scales mu lying much deeper in the low-energy regime. We are able to reach 1/mu ~ 0.9 fm, where a significant deviation from its perturbative evolution is observed., Comment: Lattice2001(improvement), 3 pages, latex2e, 5 Postscript figures, uses epsfig, amssymb and espcrc2

The scaling of the Schr"odinger functional coupling is studied numerically and perturbatively for an SU(3) lattice gauge field coupled to an O(a) improved bosonic spinor field. This corresponds to QCD with minus two light flavours and is used as a numerically less costly test case for real QCD. A suitable algorithm is developed, and the influence of the matter fields on the continuum limit and the lattice artefacts are studied in detail., Comment: 24 pages, 7 figures, small textual changes