Dominic Schuh

Dominic Schuh

Auto-generated profile, based on publicly available information

Dominic Schuh

University of Bonn

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 October 2023 in Physical Review Letters
Verified
Journal Article
Open access
PDF file
MHMartin Hoferichter
Martin Hoferichter
GCGilberto Colangelo
Gilberto Colangelo
BHBai-Long Hoid
Bai-Long Hoid
BKBastian Kubis
Bastian Kubis
JEJacobo Ruiz de Elvira
Jacobo Ruiz de Elvira
PSPeter Stoffer
Peter Stoffer
Puzzles in the determination of the hadronic-vacuum-polarization contribution currently impede a conclusive interpretation of the precision measurement of the anomalous magnetic moment of the muon at the Fermilab experiment. One such puzzle concerns tensions between evaluations in lattice QCD and using e+e−→hadrons cross-section data. In lattice QCD, the dominant isospin-symmetric part and isospin-breaking (IB) corrections are calculated separately, with very different systematic effects. Identifying these two pieces in a data-driven approach provides an opportunity to compare them individually and trace back the source of the discrepancy. Here, we estimate the IB component of the lattice-QCD calculations from phenomenology, based on a comprehensive study of exclusive contributions that can be enhanced via infrared singularities, threshold effects, or hadronic resonances, including, for the first time, in the e+e−→3π channel. We observe sizable cancellations among different channels, with a sum that even suggests a slightly larger result for the QED correction than obtained in lattice QCD. We conclude that the tensions between lattice QCD and e+e− data therefore cannot be explained by the IB contributions in the lattice-QCD calculations.

30 August 2023 in Journal of High Energy Physics
Verified
Journal Article
Open access
PDF file
MHMartin Hoferichter
Martin Hoferichter
BHBai-Long Hoid
Bai-Long Hoid
BKBastian Kubis
Bastian Kubis
DSDominic Schuh
Dominic Schuh
Isospin-breaking (IB) effects are required for an evaluation of hadronic vacuum polarization at subpercent precision. While the dominant contributions arise from the e+e−→ π+π− channel, also IB in the subleading channels can become relevant for a detailed understanding, e.g., of the comparison to lattice QCD. Here, we provide such an analysis for e+e−→ 3π by extending our dispersive description of the process, including estimates of final-state radiation (FSR) and ρ–ω mixing. In particular, we develop a formalism to capture the leading infrared-enhanced effects in terms of a correction factor η3π that generalizes the analog treatment of virtual and final-state photons in the 2π case. The global fit to the e+e−→ 3π data base, subject to constraints from analyticity, unitarity, and the chiral anomaly, gives $$ {\left.{a}_{\mu}^{3\pi}\right|}_{\le 1.8\ \textrm{GeV}}=45.91(53)\times {10}^{-10} $$ a μ 3 π ≤ 1.8 GeV = 45.91 53 × 10 − 10 for the total 3π contribution to the anomalous magnetic moment of the muon, of which $$ {a}_{\mu}^{\textrm{FSR}}\left[3\pi \right]=0.51(1)\times {10}^{-10} $$ a μ FSR 3 π = 0.51 1 × 10 − 10 and $$ {a}_{\mu}^{\rho -\omega}\left[3\pi \right]=-2.68(70)\times {10}^{-10} $$ a μ ρ − ω 3 π = − 2.68 70 × 10 − 10 can be ascribed to IB. We argue that the resulting cancellation with ρ–ω mixing in e+e−→ 2π can be understood from a narrow-resonance picture, and provide updated values for the vacuum-polarization-subtracted vector-meson parameters Mω = 782.70(3) MeV, Mϕ = 1019.21(2) MeV, Γω = 8.71(3) MeV, and Γϕ = 4.27(1) MeV.

10 February 2025
Verified
Conference Paper
Open access
PDF file
DSDominic Schuh
Dominic Schuh
JKJanik Kreit
Janik Kreit
EBEvan Berkowitz
Evan Berkowitz
LFLena Funcke
Lena Funcke
TLThomas Luu
Thomas Luu
MRMarcel Rodekamp
Marcel Rodekamp
Generative models, particularly normalizing flows, have shown exceptional performance in learning probability distributions across various domains of physics, including statistical mechanics, collider physics, and lattice field theory. In the context of lattice field theory, normalizing flows have been successfully applied to accurately learn the Boltzmann distribution, enabling a range of tasks such as direct estimation of thermodynamic observables and sampling independent and identically distributed (i.i.d.) configurations. In this work, we present a proof-of-concept demonstration that normalizing flows can be used to learn the Boltzmann distribution for the Hubbard model. This model is widely employed to study the electronic structure of graphene and other carbon nanomaterials. State-of-the-art numerical simulations of the Hubbard model, such as those based on Hybrid Monte Carlo (HMC) methods, often suffer from ergodicity issues, potentially leading to biased estimates of physical observables. Our numerical experiments demonstrate that leveraging i.i.d. sampling from the normalizing flow effectively addresses these issues.

25 March 2026 in Physical Review Research
Verified
Journal Article
Open access
DSDominic Schuh
Dominic Schuh
JKJanik Kreit
Janik Kreit
EBEvan Berkowitz
Evan Berkowitz
LFLena Funcke
Lena Funcke
TLThomas Luu
Thomas Luu
MRMarcel Rodekamp
Marcel Rodekamp
We present the first proof of principle that normalizing flows can accurately learn the Boltzmann distribution of the fermionic Hubbard model—a key framework for describing the electronic structure of graphene and related materials. State-of-the-art methods like hybrid Monte Carlo often suffer from ergodicity issues near the time-continuum limit, leading to biased estimates. Leveraging symmetry-aware architectures as well as independent and identically distributed sampling, our approach resolves these issues and achieves significant speedups over traditional methods.

30 August 2023 in arXiv
Verified
Preprint
Open access
PDF file
MHMartin Hoferichter
Martin Hoferichter
BHBai-Long Hoid
Bai-Long Hoid
BKBastian Kubis
Bastian Kubis
DSDominic Schuh
Dominic Schuh
Isospin-breaking (IB) effects are required for an evaluation of hadronic vacuum polarization at subpercent precision. While the dominant contributions arise from the $e^+e^-\toπ^+π^-$ channel, also IB in the subleading channels can become relevant for a detailed understanding, e.g., of the comparison to lattice QCD. Here, we provide such an analysis for $e^+e^-\to 3π$ by extending our dispersive description of the process, including estimates of final-state radiation (FSR) and $ρ$-$ω$ mixing. In particular, we develop a formalism to capture the leading infrared-enhanced effects in terms of a correction factor $η_{3π}$ that generalizes the analog treatment of virtual and final-state photons in the $2π$ case. The global fit to the $e^+e^-\to 3π$ data base, subject to constraints from analyticity, unitarity, and the chiral anomaly, gives $a_μ^{3π}|_{\leq 1.8\,\text{GeV}}=45.91(53)\times 10^{-10}$ for the total $3π$ contribution to the anomalous magnetic moment of the muon, of which $a_μ^\text{FSR}[3π]=0.51(1)\times 10^{-10}$ and $a_μ^{ρ\text{-}ω}[3π]=-2.68(70)\times 10^{-10}$ can be ascribed to IB. We argue that the resulting cancellation with $ρ$-$ω$ mixing in $e^+e^-\to 2π$ can be understood from a narrow-resonance picture, and provide updated values for the vacuum-polarization-subtracted vector-meson parameters $M_ω=782.70(3)\,\text{MeV}$, $M_φ=1019.21(2)\,\text{MeV}$, $Γ_ω=8.71(3)\,\text{MeV}$, and $Γ_φ=4.27(1)\,\text{MeV}$.

20 June 2025 in arXiv
Verified
Preprint
Open access
PDF file
DSDominic Schuh
Dominic Schuh
JKJanik Kreit
Janik Kreit
EBEvan Berkowitz
Evan Berkowitz
LFLena Funcke
Lena Funcke
TLThomas Luu
Thomas Luu
MRMarcel Rodekamp
Marcel Rodekamp
We present the first proof of principle that normalizing flows can accurately learn the Boltzmann distribution of the fermionic Hubbard model - a key framework for describing the electronic structure of graphene and related materials. State-of-the-art methods like Hybrid Monte Carlo often suffer from ergodicity issues near the time-continuum limit, leading to biased estimates. Leveraging symmetry-aware architectures as well as independent and identically distributed sampling, our approach resolves these issues and achieves significant speed-ups over traditional methods.