arXiv subjects

David I. Müller

Publications and source records attributed to David I. Müller.

At least 19 recordsLinked to original sources

arXiv2026

Generalizable Equivariant Diffusion Models for Non-Abelian Lattice Gauge Theory

We demonstrate that gauge equivariant diffusion models can accurately model the physics of non-Abelian lattice gauge theory using the Metropolis-adjusted annealed Langevin algorithm (MAALA), as exemplified by computations in two-dimensional U(2) and SU(2) gauge theories. Our network architecture is based on lattice gauge equivariant convolutional neural networks (L-CNNs), which respect local and global symmetries on the lattice. Models are trained on a single ensemble generated using a traditional Monte Carlo method. By studying Wilson loops of various size as well as the topological susceptibility, we find that the diffusion approach generalizes remarkably well to larger inverse couplings and lattice sizes with negligible loss of accuracy while retaining moderately high acceptance rates.

hep-lat↗

arXiv2026

Machine-learned RG-improved gauge actions and classically perfect gradient flows

Extracting continuum properties of quantum field theories from discretized spacetime is challenging due to lattice artifacts. Renormalization-group (RG)-improved lattice actions can preserve continuum properties, but are in general difficult to parameterize. Machine learning (ML) with gauge-equivariant convolutional neural networks provides a way to efficiently describe such actions. We test a machine-learned RG-improved lattice gauge action, the classically perfect fixed-point (FP) action, for four-dimensional SU(3) gauge theory through Monte Carlo simulations. We establish that the gradient flow of the FP action is free of tree-level discretization effects to all orders in the lattice spacing, making it classically perfect. This allows us to test the quality of improvement of the FP action, without introducing additional artifacts. We find that discretization effects in gradient-flow observables are highly suppressed and less than 1% up to lattice spacings of 0.14 fm, allowing continuum physics to be extracted from coarse lattices. The quality of improvement achieved motivates the use of the FP action in future gauge theory studies. The advantages of ML-based parameterizations also highlight the possibility of realizing quantum perfect actions in lattice gauge theory.

hep-lat↗

arXiv2025

Heavy flavor angular correlations as probes of the glasma

We study the effect of the glasma fields, formed in the early stage of heavy-ion collisions, on the transport of $Q\bar{Q}$ pairs produced back-to-back. We find that for pairs with moderate initial transverse momentum $p_T$ evolving in glasma fields with sufficiently large saturation momentum $Q_s$, the azimuthal correlation $C({Δϕ})$ is quickly affected. The decorrelation widths $σ_{Δϕ}$ during the glasma and Quark Gluon Plasma (QGP) phases are comparable.

hep-ph↗

arXiv2025

The (3+1)D structure of the dilute Glasma

We study the (3+1)D structure of the Glasma in the dilute approximation, which allows us to describe the longitudinal dynamics that arise from the three-dimensional nuclear structure. We employ a nuclear model with tunable longitudinal and transverse fluctuation scales that generalizes the McLerran-Venugopalan model. We discuss the longitudinal profiles of the energy-momentum tensor and the transverse structure of the local rest frame energy density.

hep-ph↗

arXiv2025

Real-time correlators in 3+1D thermal lattice gauge theory

We present the first direct ab-initio computation of unequal-time correlation functions in non-Abelian lattice gauge theory. We demonstrate non-trivial consistency relations among correlators, time-translation invariance, and agreement with Monte-Carlo results for thermal equilibrium in 3+1 dimensions by employing our stabilized complex Langevin method. Our work sets the stage to extract real-time observables, relevant to quark-gluon plasma physics within a first-principles real-time framework.

hep-lat↗

arXiv2025

Lefschetz thimble-inspired weight regularizations for complex Langevin simulations

Complex Langevin (CL) is a computational method to circumvent the numerical sign problem with applications in finite-density quantum chromodynamics and the real-time dynamics of quantum field theories. It has long been known that, depending on the simulated system, CL does not always converge correctly. In this work, we provide numerical evidence that the success or failure of the complex Langevin method is deeply tied to the Lefschetz thimble structure of the simulated system. This is demonstrated by constructing weight function regularizations that deform the thimbles of systems with compact domains. Our results indicate that CL converges correctly when the regularized system exhibits a single relevant compact thimble. We introduce a bias correction to retrieve the values of the original theory for parameter sets where a direct complex Langevin approach fails. The effectiveness of this method is illustrated using several toy models, including the cosine model and the SU(2) and SU(3) Polyakov chains. Finally, we discuss the opportunities and limitations of this regularization approach for lattice field theories.

hep-lat↗

arXiv2025

Limiting fragmentation in the dilute Glasma

We discuss the local longitudinal scaling behavior of the dilute Glasma. We gain insight into how the fragmentation region is dominated by the longitudinal structure of one of the two colliding nuclei in heavy-ion collisions and study the effect of the correlation scales of the nuclear color charge distributions. We compare with results from the full dilute Glasma framework and analyze the rapidity limits of the fragmentation region. From our findings, it follows that all scalar observables that can be constructed out of the dilute Glasma field strength tensor show limiting fragmentation behavior and the field strength tensor itself transforms locally via Lorentz boosts when changing the collision energy.

hep-ph↗

arXiv2025

Charged particle multiplicity in pp-collisions from the dilute Glasma

Proton-proton collisions are studied in the dilute Glasma framework. Compared to experimental multiplicity distributions, the dilute Glasma underestimates large multiplicity events. We show how event-by-event fluctuations of the saturation momentum Qs can help repair the multiplicity distribution. Furthermore, we discuss a hot spot model for protons and also find improvements in the multiplicity histograms.

hep-ph↗

arXiv2025

HMC and gradient flow with machine-learned classically perfect fixed-point actions

Fixed-point (FP) lattice actions are classically perfect, i.e., they have continuum classical properties unaffected by discretization effects and are expected to have suppressed lattice artifacts at weak coupling. Therefore they provide a possible way to extract continuum physics with coarser lattices, allowing to circumvent problems with critical slowing down and topological freezing towards the continuum limit. We use machine-learning methods to parameterize a FP action for four-dimensional SU(3) gauge theory using lattice gauge-covariant convolutional neural networks. The large operator space allows us to find superior parameterizations compared to previous studies and we show how such actions can be efficiently simulated with the Hybrid Monte Carlo algorithm. Furthermore, we argue that FP lattice actions can be used to define a classically perfect gradient flow without any lattice artifacts at tree level. We present initial results for scaling of the gradient flow with the FP action.

hep-lat↗

arXiv2024

Designing weight regularizations based on Lefschetz thimbles to stabilize complex Langevin

The complex Langevin (CL) method shows significant potential in addressing the numerical sign problem. Nonetheless, it often produces incorrect results when used without any stabilization techniques. Leveraging insights from previous research that links Lefschetz thimbles and CL, we explore a strategy to regularize the CL method to address this issue of incorrect convergence. Specifically, we implement weight regularizations inspired by the associated Lefschetz thimble structure and correct the bias to retrieve the correct results of the original theory. We demonstrate the effectiveness of this approach by solving the SU(N) Polyakov chain model and various scalar models, including the cosine model and the one-link model, across a broad range of couplings where the CL method previously failed. We also discuss the potential application of these insights to gauge theories in practical scenarios.

hep-lat↗

arXiv2024

Machine learning a fixed point action for SU(3) gauge theory with a gauge equivariant convolutional neural network

Fixed point lattice actions are designed to have continuum classical properties unaffected by discretization effects and reduced lattice artifacts at the quantum level. They provide a possible way to extract continuum physics with coarser lattices, thereby allowing one to circumvent problems with critical slowing down and topological freezing toward the continuum limit. A crucial ingredient for practical applications is to find an accurate and compact parametrization of a fixed point action, since many of its properties are only implicitly defined. Here we use machine learning methods to revisit the question of how to parametrize fixed point actions. In particular, we obtain a fixed point action for four-dimensional SU(3) gauge theory using convolutional neural networks with exact gauge invariance. The large operator space allows us to find superior parametrizations compared to previous studies, a necessary first step for future Monte Carlo simulations and scaling studies.

hep-lat↗

arXiv2024

Energy-momentum tensor of the dilute (3+1)D Glasma

We present a succinct formulation of the energy-momentum tensor of the Glasma characterizing the initial color fields in relativistic heavy-ion collisions in the Color Glass Condensate effective theory. We derive concise expressions for the (3+1)D dynamical evolution of symmetric nuclear collisions in the weak field approximation employing a generalized McLerran-Venugopalan model with non-trivial longitudinal correlations. Utilizing Monte Carlo integration, we calculate in unprecedented detail non-trivial rapidity profiles of early-time observables at RHIC and LHC energies, including transverse energy densities and eccentricities. For our setup with broken boost invariance, we carefully discuss the placement of the origin of the Milne frame and interpret the components of the energy-momentum tensor. We find longitudinal flow that deviates from standard Bjorken flow in the (3+1)D case and provide a geometric interpretation of this effect. Furthermore, we observe a universal shape in the flanks of the rapidity profiles regardless of collision energy and predict that limiting fragmentation should also hold at LHC energies.

hep-ph↗

arXiv2024

Advancing real-time Yang-Mills: towards real-time observables from first principles

The complex Langevin (CL) method shows great promise in enabling the calculation of observables for theories with complex actions. Nevertheless, real-time quantum field theories have remained largely unsolved due to the particular severity of the sign problem. In this contribution, we discuss our recent progress in applying CL to a thermal SU(2) Yang-Mills theory on a 3+1 dimensional lattice. We present our anisotropic kernel that stabilizes the CL approach for real times longer than the inverse temperature - a first for Yang-Mills theory. We provide explicit evidence of reproducing symmetries and relations among different types of propagators when the complex time path approaches the Schwinger-Keldysh contour. This method paves the way for calculating transport coefficients and other real-time observables from first principles.

hep-lat↗

arXiv2023

Highly anisotropic lattices for Yang-Mills theory

In this conference proceeding, we investigate the physical anisotropy in terms of the temporal and spatial lattice spacings in relation to the bare parameters of SU(2) pure gauge theory using Wilson gradient flow. Anisotropic lattices have a wide range of applications, from thermodynamic calculations in QCD to very recent real-time simulations using the complex Langevin method. We find an almost linear relationship between the bare and renormalized anisotropy. Using a parametrization that includes nonlinear effects and was earlier proposed for SU(3) theory, we obtain a good description of the coupling dependence of the anisotropy with only two fitting parameters. Our observation of an approximately linear relationship and this parametrization should strongly reduce the computational effort of anisotropic lattice calculations in the future.

hep-lat↗

arXiv2023

Fixed point actions from convolutional neural networks

Lattice gauge-equivariant convolutional neural networks (L-CNNs) can be used to form arbitrarily shaped Wilson loops and can approximate any gauge-covariant or gauge-invariant function on the lattice. Here we use L-CNNs to describe fixed point (FP) actions which are based on renormalization group transformations. FP actions are classically perfect, i.e., they have no lattice artifacts on classical gauge-field configurations satisfying the equations of motion, and therefore possess scale invariant instanton solutions. FP actions are tree-level Symanzik-improved to all orders in the lattice spacing and can produce physical predictions with very small lattice artifacts even on coarse lattices. We find that L-CNNs are much more accurate at parametrizing the FP action compared to older approaches. They may therefore provide a way to circumvent critical slowing down and topological freezing towards the continuum limit.

hep-lat↗

arXiv2023

Stabilizing complex Langevin for real-time gauge theories with an anisotropic kernel

The complex Langevin (CL) method is a promising approach to overcome the sign problem that occurs in real-time formulations of quantum field theories. Using the Schwinger-Keldysh formalism, we study SU($N_c$) gauge theories with CL. We observe that current stabilization techniques are insufficient to obtain correct results. Therefore, we revise the discretization of the CL equations on complex time contours, find a time reflection symmetric formulation and introduce a novel anisotropic kernel that enables CL simulations on discretized complex time paths. Applying it to SU(2) Yang-Mills theory in 3+1 dimensions, we obtain unprecedentedly stable results that we validate using additional observables and that can be systematically improved. For the first time, we are able to simulate non-Abelian gauge theory on time contours whose real-time extent exceeds its inverse temperature. Thus, our approach may pave the way towards an ab-initio real-time framework of QCD in and out of equilibrium with a potentially large impact on the phenomenology of heavy-ion collisions.

hep-lat↗

arXiv2023

Geometrical aspects of lattice gauge equivariant convolutional neural networks

Lattice gauge equivariant convolutional neural networks (L-CNNs) are a framework for convolutional neural networks that can be applied to non-Abelian lattice gauge theories without violating gauge symmetry. We demonstrate how L-CNNs can be equipped with global group equivariance. This allows us to extend the formulation to be equivariant not just under translations but under global lattice symmetries such as rotations and reflections. Additionally, we provide a geometric formulation of L-CNNs and show how convolutions in L-CNNs arise as a special case of gauge equivariant neural networks on SU($N$) principal bundles.

hep-lat↗

arXiv2022

Studying the 3+1D structure of the Glasma using the weak field approximation

We extend the weak field approximation for the Glasma beyond the boost-invariant approximation, which allows us to compute rapidity-dependent observables in the early stages of heavy-ion collisions. We show that in the limit of small fields, the weak field approximation agrees quantitatively with non-perturbative lattice simulations. Furthermore, we demonstrate that the rapidity profile of the transverse pressure is determined by longitudinal color correlations within the colliding nuclei.

hep-ph↗