
This is the report of the Computing Frontier working group on Lattice Field
Theory prepared for the proceedings of the 2013 Community Summer Study
("Snowmass"). We present the future computing needs and plans of the U.S.
lattice gauge theory community and argue that continued support of the U.S.
(and worldwide) latticeQCD effort is essential to fully capitalize on the
enormous investment in the highenergy physics experimental program. We first
summarize the dramatic progress of numerical latticeQCD simulations in the
past decade, with some emphasis on calculations carried out under the auspices
of the U.S. LatticeQCD Collaboration, and describe a broad program of
latticeQCD calculations that will be relevant for future experiments at the
intensity and energy frontiers. We then present details of the computational
hardware and software resources needed to undertake these calculations.

We show that lattice regularization of noncommutative field theories can be
used to study nonperturbative vacuum phases. Specifically we provide evidence
for the existence of a striped phase in twodimensional noncommutative scalar
field theory

We describe a Fourier Accelerated Hybrid Monte Carlo algorithm suitable for
dynamical fermion simulations of nongauge models. We test the algorithm in
supersymmetric quantum mechanics viewed as a onedimensional Euclidean lattice
field theory. We find dramatic reductions in the autocorrelation time of the
algorithm in comparison to standard HMC.

Starting from a simple discrete model which exhibits a supersymmetric
invariance we construct a local, interacting, twodimensional Euclidean lattice
theory which also admits an exact supersymmetry. This model is shown to
correspond to the WessZumino model with extended N=2 supersymmetry in the
continuum. We have performed dynamical fermion simulations to check the
spectrum and supersymmetric Ward identities and find good agreement with
theory.

We verify that summing 2D DT geometries correctly reproduces the Polyakov
action for the conformal mode, including all ghost contributions, at large
volumes. The Gaussian action is reproduced even for central charges greater
than one lending strong support to the hypothesis that the space of all
possible dyamical triangulations approximates well the space of physically
distinct metrics independent of the precise nature of the matter coupling.

We verify that summing 2D DT geometries correctly reproduces the Polyakov
action for the conformal mode, including all ghost contributions, at large
volumes. The Gaussian action is reproduced even for c=10, well into the
branched polymer phase, which confirms the expectation that the DT measure is
indeed correct in this regime as well.

We describe the statistical behavior of anisotropic crystalline membranes. In
particular we give the phase diagram and critical exponents for phantom
membranes and discuss the generalization to selfavoiding membranes.

We review the derivation of the Liouville action in 2DQG via the trace
anomaly and emphasize how a similar approach can be used to derive an effective
action describing the long wavelength dynamics of the conformal factor in 4D.
In 2D we describe how to make an explicit connection between dynamical
triangulations and this continuum theory, and present results which confirm the
equivalance of the two approaches. By reconstructing a lattice conformal mode
from DT simulations it should be possible to test this equivalence in 4D also.

We show that the phase transition previously observed in dynamical
triangulation models of quantum gravity can be understood as being due to the
creation of a singular link. The transition between singular and nonsingular
geometries as the gravitational coupling is varied appears to be first order.

We consider two issues in the DT model of quantum gravity. First, it is shown
that the triangulation space for D>3 is dominated by triangulations containing
a single singular (D3)simplex composed of vertices with divergent dual
volumes. Second we study the ergodicity of current simulation algorithms.
Results from runs conducted close to the phase transition of the
fourdimensional theory are shown. We see no strong indications of ergodicity
br eaking in the simulation and our data support recent claims that the
transition is most probably first order. Furthermore, we show that the critical
properties of the system are determined by the dynamics of remnant singular
vertices.

The statistical mechanics of flexible twodimensional surfaces (membranes)
appears in a wide variety of physical settings. In this talk we discuss the
simplest case of fixedconnectivity surfaces. We first review the current
theoretical understanding of the remarkable flat phase of such membranes. We
then summarize the results of a recent large scale Monte Carlo simulation of
the simplest conceivable discrete realization of this system \cite{BCFTA}. We
verify the existence of longrange order, determine the associated critical
exponents of the flat phase and compare the results to the predictions of
various theoretical models.

We present the results of a highstatistics Monte Carlo simulation of a
phantom crystalline (fixedconnectivity) membrane with free boundary. We verify
the existence of a flat phase by examining lattices of size up to $128^2$. The
Hamiltonian of the model is the sum of a simple spring pair potential, with no
hardcore repulsion, and bending energy. The only free parameter is the the
bending rigidity $\kappa$. Inplane elastic constants are not explicitly
introduced. We obtain the remarkable result that this simple model dynamically
generates the elastic constants required to stabilise the flat phase. We
present measurements of the size (Flory) exponent $\nu$ and the roughness
exponent $\zeta$. We also determine the critical exponents $\eta$ and $\eta_u$
describing the scale dependence of the bending rigidity ($\kappa(q) \sim
q^{\eta}$) and the induced elastic constants ($\lambda(q) \sim \mu(q) \sim
q^{\eta_u}$). At bending rigidity $\kappa = 1.1$, we find $\nu = 0.95(5)$
(Hausdorff dimension $d_H = 2/\nu = 2.1(1)$), $\zeta = 0.64(2)$ and $\eta_u =
0.50(1)$. These results are consistent with the scaling relation $\zeta =
(2+\eta_u)/4$. The additional scaling relation $\eta = 2(1\zeta)$ implies
$\eta = 0.72(4)$. A direct measurement of $\eta$ from the powerlaw decay of
the normalnormal correlation function yields $\eta \approx 0.6$ on the $128^2$
lattice.

By a sequence of numerical experiments we demonstrate that generic
triangulations of the $D$sphere for $D>3$ contain one {\it singular}
$(D3)$simplex. The mean number of elementary $D$simplices sharing this
simplex increases with the volume of the triangulation according to a simple
power law. The lower dimension subsimplices associated with this
$(D3)$simplex also show a singular behaviour. Possible consequences for the
DT model of fourdimensional quantum gravity are discussed.

We review the status of different approaches to lattice quantum gravity
indicating the successes and problems of each. Recent developments within the
dynamical triangulation formulation are then described. Plenary talk at LATTICE
95 July 1115, Melbourne, Australia.

We propose a new realspace renormalization group transformation for
dynamical triangulations. It is shown to preserve geometrical exponents such as
the string susceptibility and Hausdorff dimension. We furthermore show evidence
for a fixed point structure both in pure gravity and gravity coupled to a
critical Ising system. In the latter case we are able to extract estimates for
the gravitationally dressed exponents which agree to within 23% of the KPZ
formula.

We examine the scaling of geodesic correlation functions in twodimensional
gravity and in spin systems coupled to gravity. The numerical data support the
scaling hypothesis and indicate that the quantum geometry develops a
nonperturbative length scale. The existence of this length scale allows us to
extract a Hausdorff dimension. In the case of pure gravity we find d_H approx.
3.8, in support of recent theoretical calculations that d_H = 4. We also
discuss the backreaction of matter on the geometry.

We present numerical results supporting the existence of an exponential bound
in the dynamical triangulation model of threedimensional quantum gravity.Both
the critical coupling and various other quantities show a slow power law
approach to the infinite volume limit.

Recent models for discrete euclidean quantum gravity incorporate a sum over
simplicial triangulations. We describe an algorithm for simulating such models
in general dimensions. As illustration we show results from simulations in four
dimensions

We have studied a model which has been proposed as a regularisation for four
dimensional quantum gravity. The partition function is constructed by
performing a weighted sum over all triangulations of the four sphere. Using
numerical simulation we find that the number of such triangulations containing
$V$ simplices grows faster than exponentially with $V$. This property ensures
that the model has no thermodynamic limit.

We present the results of a high statistics Monte Carlo study of a model for
four dimensional euclidean quantum gravity based on summing over
triangulations. We show evidence for two phases; in one there is a logarithmic
scaling on the mean linear extent with volume, whilst the other exhibits power
law behaviour with exponent 1/2. We are able to extract a finite size scaling
exponent governing the growth of the susceptibility peak

We study the XY model on a lattice with fluctuating connectivity. The
expectation is that at an appropriate critical point such a system corresponds
to a compactified boson coupled to 2d quantum gravity. Our simulations focus,
in particular, on the important topological features of the system. The results
lend strong support to the two phase structure predicted on the basis of
analytical calculations. A careful finite size scaling analysis yields
estimates for the critical exponents in the low temperature phase.