betaFeedback

WorksPhysics and Astronomy

Nonperturbative functional renormalization group for Higgs-singlet models with physics-informed neural networks

Abstract

We develop a nonperturbative functional renormalization group framework within the LPA' to solve the Wetterich flow equation for the Z2Z_2-symmetric real singlet extension of the Standard Model at finite temperature, without a low-order polynomial truncation of the loop corrections to the effective potential, using a physics-informed neural network (PINN) representation. In contrast to conventional truncations based on low-order field expansions, our hybrid tree-level-plus-neural-network ansatz yields a continuous, mesh-free description of the effective potential over the full field and scale range. As a proof of concept, we apply the framework to the finite-temperature effective potential relevant to the electroweak phase transition: one-dimensional field-space slices at two benchmark temperatures, and a two-dimensional reconstruction at T=100 GeV yielding a two-step bounce action. The flow is implemented in a multi-domain setup in scale and field space with derivative matching conditions ensuring smoothness and numerical stability. Gauge and Yukawa sectors are incorporated via independently computed perturbative two-loop running couplings, and anomalous dimensions are included in the flow at the LPA' level. We benchmark the network against resummed perturbation theory and against a grid-based relaxation solver of the same equation. Finally, we introduce a soft consistency constraint that keeps the solution close, in sign and order of magnitude, to perturbation theory across the field-space domain. We find this constraint necessary, rather than merely helpful, for selecting a physically sensible solution of the flow equation. The converged result nevertheless retains a residual dependence on this guidance – through hand-tuned weights and an analytic thermal target – which we identify as the central open problem for mesh-free FRG treatments of this kind.

Introduction

The Standard Model (SM) of particle physics, completed by the discovery of the Higgs boson [1, 2], describes all observed collider phenomena with remarkable accuracy and remains consistent with essentially every precision test performed to date.

Despite this success, the SM cannot be the complete theory of nature. It does not account for neutrino masses, dark matter, or the origin of the observed cosmic matter–antimatter asymmetry, leaving several central problems of cosmology and particle physics unresolved.

Among these, the baryon asymmetry of the universe (BAU) is a particularly sharp challenge: the SM cannot generate the observed asymmetry through the Sakharov conditions in sufficient quantity [3, 4, 5, 6]. This motivates extending the scalar sector of the SM in ways that can support a viable baryogenesis mechanism while remaining compatible with current collider and dark matter constraints.

Electroweak baryogenesis (EWBG) offers an attractive mechanism to generate the BAU at the electroweak scale, but it requires two separate ingredients absent from the SM: a strongly first-order EWPT and new sources of CP violation. In the SM itself neither is realized – lattice studies show that the electroweak transition is a smooth crossover rather than a first-order transition [7], and the CP violation of the CKM matrix is far too small to seed the observed asymmetry – so EWBG is only a viable option once the scalar sector is extended. Extending the SM with an additional scalar, most simply a real singlet, is one of the best-studied ways to restore a strongly first-order EWPT: the extra scalar can modify the shape of the finite-temperature potential enough to generate a barrier between the symmetric and broken phases, and such singlet extensions have been extensively analyzed in connection with dark matter, vacuum stability, and the EWPT, both for the real, Z2Z_2-symmetric case that we adopt below [8, 9, 10, 11, 12] and, once the singlet is complexified or coupled with additional CP-odd interactions, as a source of the CP violation needed to complete EWBG [13, 14]. Reliable computations of the finite-temperature effective potential are essential for establishing whether a given extension actually achieves a strongly first-order transition, but conventional perturbative techniques suffer from severe IR sensitivity and resummation issues near the critical temperature of a first-order EWPT.

The functional renormalization group (FRG), formulated via the Wetterich equation, provides a non-perturbative tool which smoothly interpolates between the bare UV action and the full IR effective action, and is well suited to resolving these IR issues. However, in scalar extensions of the SM the effective potential depends on multiple field invariants, leading to a nonlinear PDE in more than one variable, which is technically demanding to solve by standard finite-difference or pseudo-spectral methods. Most functional renormalization group studies of scalar extensions of the Standard Model rely on low-order polynomial truncations of the effective potential or grid-based discretizations in field space.

Here we consider the real singlet scalar extension of the SM with Z2Z_2 symmetry as a concrete setting for this FRG treatment. Being Z2Z_2-symmetric and real, the singlet introduces no new source of CP violation, so the model by itself cannot complete EWBG; we take the strongly first-order transition as the first step to be established quantitatively, using this model as a controlled testing ground for the finite-temperature effective-potential methodology, and leave the additional CP-violating sector required to generate the BAU to future extensions of the model.

Physics-informed neural networks (PINNs) [15] parametrize the unknown function as a neural network and minimize the PDE residual via automatic differentiation. They are mesh-free and naturally handle multi-dimensional domains, making them a natural candidate for the multi-field FRG flow equations that are otherwise costly to solve. Applying this idea to the FRG flow equation itself is a recent development: a genuinely mesh-free treatment, in which a neural network represents the field dependence continuously and the flow equation is enforced through automatic differentiation of the network output, has been demonstrated for a single-field O(N)O(N)-symmetric scalar theory [16], and a related neural-network parametrization of the field- and scale-dependent potential for the O(N)O(N) model at finite temperature has also been proposed, though there the network is trained against a finite-difference-discretized residual evaluated on a fixed field-and-RG-time grid rather than a continuous PDE residual [17]. Extending a genuinely continuous, autodiff-based PINN treatment to a multi-field, finite-temperature effective potential of a phenomenologically motivated BSM extension – as required for a first-order EWPT analysis – remains unexplored. In this paper we apply PINNs to the FRG equation of the Z2Z_2-symmetric singlet extension of the SM with finite-temperature contributions and SM gauge/Yukawa loops. Our aim here is methodological: to establish the end-to-end pipeline – from UV couplings, through the mesh-free flow, to a finite-temperature effective potential and a bounce action – as a proof of concept, rather than to deliver a full temperature scan or a converged determination of the transition strength at this benchmark.

The remainder of this paper is organized as follows. Sec. 2 sets up the Wetterich equation for the Z2\mathbb{Z}_2 singlet extension and its dimensionless LPA′' truncation. Sec. 3 reviews the resulting two-step electroweak phase transition and the critical and nucleation temperatures used to assess a strongly first-order transition. Sec. 4 fixes the benchmark point and input parameters used throughout, including the one-loop Coleman–Weinberg matching at kIRk_{\mathrm{IR}} and the two-loop RGE running to kUVk_{\mathrm{UV}}, and introduces a grid-based Newton–Krylov relaxation solver as a benchmark for the flow equation. It first checks the solver on a singlet-free reduction of the model – the simplest, single-field version of the problem – and then on the full two-field system, so as to test whether the observed convergence failure originates in the Higgs–singlet mass-eigenvalue mixing or is already present in the single-field loop-level flow equation. Sec. 5 describes the PINN architecture, decomposition of the potential, loss function, and multi-block training scheme used to solve the same flow equation. Sec. 6 presents the resulting PINN solutions, again starting from the singlet-free reduction before the full system, compares them with the perturbative and relaxation-based benchmarks, and discusses the practical role and limitations of the soft perturbative-consistency penalty used in the PINN loss. Sec. 7 summarizes our findings and outlines directions for future work, including the CP-violating extensions needed to complete EWBG.

Wetterich equation for the Z2\mathbb{Z}_2 singlet extension

Model and truncation

We consider the scalar sector of the Standard Model extended by a real singlet scalar SS with a Z2\mathbb{Z}_2 symmetry,

S→−S.(1)S \to-S. \tag*{(1)}

In this subsection, we keep only scalar fluctuations and neglect gauge and Yukawa contributions. At the UV scale k=Λk = \Lambda, the Z2\mathbb{Z}_2-symmetric scalar potential is

VΛ(H,S)=μH2∣H∣2+λH∣H∣4+12μS2S2+14λSS4+12λHS∣H∣2S2.(2)V_{\Lambda}(H,S) = \mu_H^2 |H|^2 + \lambda_H |H|^4 + \frac{1}{2}\mu_S^2 S^2 + \frac{1}{4}\lambda_S S^4 + \frac{1}{2}\lambda_{HS}|H|^2S^2. \tag*{(2)}

Introducing the field invariants

ρ~≡H†H,σ~≡12S2,(3)\tilde{\rho} \equiv H^\dagger H,\qquad\tilde{\sigma} \equiv\frac{1}{2}S^2, \tag*{(3)}

the potential can be written as

VΛ(ρ~,σ~)=μH2ρ~+λHρ~2+μS2σ~+λSσ~2+λHSρ~σ~.(4)V_{\Lambda}(\tilde{\rho},\tilde{\sigma}) = \mu_H^2\tilde{\rho} + \lambda_H\tilde{\rho}^2 + \mu_S^2\tilde{\sigma} + \lambda_S\tilde{\sigma}^2 + \lambda_{HS}\tilde{\rho}\tilde{\sigma}. \tag*{(4)}

Within the derivative expansion at LPA′' [18], we employ the ansatz

Γk[H,S]=∫x[ZH,k(∂μH)†(∂μH)+12ZS,k(∂μS)2+Vk(ρ~,σ~)].(5)\Gamma_k[H,S] = \int_x \left[ Z_{H,k}(\partial_\mu H)^\dagger(\partial_\mu H) + \frac{1}{2}Z_{S,k}(\partial_\mu S)^2 + V_k(\tilde{\rho},\tilde{\sigma}) \right]. \tag*{(5)}

where

∫x≡∫01/Tdτ∫d3x,t≡ln⁡kΛ.(6)\int_x \equiv\int_0^{1/T} d\tau\int d^3x,\qquad t \equiv\ln\frac{k}{\Lambda}. \tag*{(6)}

and the anomalous dimensions are defined by

ηH≡−∂tln⁡ZH,k,ηS≡−∂tln⁡ZS,k.(7)\eta_H \equiv-\partial_t \ln Z_{H,k},\qquad\eta_S \equiv-\partial_t \ln Z_{S,k}. \tag*{(7)}

Wetterich equation

The flow of the effective average action is governed by the Wetterich equation [19, 20],

∂tΓk=12STr⁡[(Γk(2)+Rk)−1∂tRk].(8)\partial_t\Gamma_k = \frac{1}{2}\operatorname{STr}\left[\left(\Gamma_k^{(2)}+R_k\right)^{-1}\partial_tR_k\right]. \tag*{(8)}

At finite temperature, the Euclidean zeroth component is replaced by the bosonic Matsubara frequencies,

p0→ωn=2πnT,∫dp02π→T∑n∈Z.(9)p_0 \to\omega_n = 2\pi nT,\qquad\int\frac{\mathrm{d}p_0}{2\pi}\to T\sum_{n\in\mathbb{Z}}. \tag*{(9)}

Using the Litim regulator [21],

RH,k(p)=ZH,k(k2−p2)θ(k2−p2),(10)R_{H,k}(p)=Z_{H,k}\left(k^2-p^2\right)\theta\left(k^2-p^2\right), \tag*{(10)}
RS,k(p)=ZS,k(k2−p2)θ(k2−p2),(11)R_{S,k}(p)=Z_{S,k}\left(k^2-p^2\right)\theta\left(k^2-p^2\right), \tag*{(11)}

and evaluating the flow for homogeneous background fields, the momentum-shell integral is elementary and the Matsubara sum closes in terms of a coth [22], giving the flow of the effective potential,

∂tVk(ρ~,σ~)=k412π2[31+mˉG2(ρ~,σ~)coth⁡(1+mˉG2(ρ~,σ~)2τ)+11+mˉ+2(ρ~,σ~)coth⁡(1+mˉ+2(ρ~,σ~)2τ)+11+mˉ−2(ρ~,σ~)coth⁡(1+mˉ−2(ρ~,σ~)2τ)]+k4Eext,(12)\begin{aligned} \partial_tV_k(\tilde{\rho},\tilde{\sigma})={}&\frac{k^4}{12\pi^2}\left[\frac{3}{\sqrt{1+\bar{m}_G^2(\tilde{\rho},\tilde{\sigma})}}\coth\left(\frac{\sqrt{1+\bar{m}_G^2(\tilde{\rho},\tilde{\sigma})}}{2\tau}\right)\right.\\ &+\frac{1}{\sqrt{1+\bar{m}_+^2(\tilde{\rho},\tilde{\sigma})}}\coth\left(\frac{\sqrt{1+\bar{m}_+^2(\tilde{\rho},\tilde{\sigma})}}{2\tau}\right)\\ &\left.+\frac{1}{\sqrt{1+\bar{m}_-^2(\tilde{\rho},\tilde{\sigma})}}\coth\left(\frac{\sqrt{1+\bar{m}_-^2(\tilde{\rho},\tilde{\sigma})}}{2\tau}\right)\right]+k^4E_{\mathrm{ext}}, \tag*{(12)} \end{aligned}

where τ=T/k\tau=T/k. Each bosonic mode ii in the bracket arises as

References

  1. [1]Georges Aad et al. Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC. Phys. Lett. B, 716:1–29, 2012.
  2. [2]Serguei Chatrchyan et al. Observation of a New Boson at a Mass of 125 GeV with the CMS Experiment at the LHC. Phys. Lett. B, 716:30–61, 2012.
  3. [3]A. D. Sakharov. Violation of CP Invariance, C asymmetry, and baryon asymmetry of the universe. Pisma Zh. Eksp. Teor. Fiz., 5:32–35, 1967.
  4. [4]V. A. Kuzmin, V. A. Rubakov, and M. E. Shaposhnikov. On the Anomalous Electroweak Baryon Number Nonconservation in the Early Universe. Phys. Lett. B, 155:36, 1985.
  5. [5]David E. Morrissey and Michael J. Ramsey-Musolf. Electroweak baryogenesis. New J. Phys., 14:125003, 2012.
  6. [6]Jorinde van de Vis, Jordy de Vries, and Marieke Postma. Bubble Trouble: a Review on Electroweak Baryogenesis. arXiv:2508.09989, 2025.DOI
  7. [7]K. Kajantie, M. Laine, K. Rummukainen, and M. E. Shaposhnikov. Is there a hot electroweak phase transition at m(H) larger or equal to m(W)? Phys. Rev. Lett., 77:2887–2890, 1996.
  8. [8]J. R. Espinosa and M. Quiros. The Electroweak phase transition with a singlet. Phys. Lett. B, 305:98–105, 1993.
  9. [9]Stefano Profumo, Michael J. Ramsey-Musolf, and Gabe Shaughnessy. Singlet Higgs phenomenology and the electroweak phase transition. JHEP, 08:010, 2007.DOI
  10. [10]Parsa Ghorbani. Vacuum structure and electroweak phase transition in singlet scalar dark matter. Phys. Dark Univ., 33:100861, 2021.
  11. [11]Lauri Niemi, Philipp Schicho, and Tuomas V. I. Tenkanen. Singlet-assisted electroweak phase transition at two loops. Phys. Rev. D, 103(11):115035, 2021.
  12. [12]Lauri Niemi, Michael J. Ramsey-Musolf, and Guotao Xia. Nonperturbative study of the electroweak phase transition in the real scalar singlet extended Standard Model. Phys. Rev. D, 110(11):115016, 2024.arxiv.org/abs/2405.01191
  13. [13]James M. Cline and Kimmo Kainulainen. Electroweak baryogenesis and dark matter from a singlet Higgs. JCAP, 01:012, 2013.
  14. [14]Bohdan Grzadkowski and Da Huang. Spontaneous CP-Violating Electroweak Baryogenesis and Dark Matter from a Complex Singlet Scalar. JHEP, 08:135, 2018.DOI
  15. [15]M. Raissi, P. Perdikaris, and G. E. Karniadakis. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys., 378:686–707, 2019.DOI
  16. [16]Takeru Yokota. Physics-informed neural networks for solving functional renormalization group on a lattice. Phys. Rev. B, 109(21):214205, 2024.DOI
  17. [17]Yang-yang Tan, Wei-jie Fu, Lianyi He, and Lingxiao Wang. Solving Functional Renormalization Group Equations with Neural Networks. arXiv:2603.21151, accepted for publication in Phys. Rev. D, 2026.DOI
  18. [18]N. Dupuis, L. Canet, A. Eichhorn, W. Metzner, J. M. Pawlowski, M. Tissier, and N. Wschebor. The nonperturbative functional renormalization group and its applications. Phys. Rept., 910:1–114, 2021.
  19. [19]Christof Wetterich. Exact evolution equation for the effective potential. Phys. Lett. B, 301:90–94, 1993.
  20. [20]Jurgen Berges, Nikolaos Tetradis, and Christof Wetterich. Nonperturbative renormalization flow in quantum field theory and statistical physics. Phys. Rept., 363:223–386, 2002.DOI
  21. [21]Daniel F. Litim. Optimized renormalization group flows. Phys. Rev. D, 64:105007, 2001.
  22. [22]Daniel F. Litim and Jan M. Pawlowski. Non-perturbative thermal flows and resummations. JHEP, 11:026, 2006.DOI
  23. [23]Lohan Sartore and Ingo Schienbein. PyR@TE 3. Comput. Phys. Commun., 261:107819, 2021.DOI
  24. [24]Sidney R. Coleman. The Fate of the False Vacuum. 1. Semiclassical Theory. Phys. Rev. D, 15:929–936, 1977.
  25. [25]Andrei D. Linde. Decay of the False Vacuum at Finite Temperature. Nucl. Phys. B, 216:421, 1983.
  26. [26]Carroll L. Wainwright. CosmoTransitions: Computing Cosmological Phase Transition Temperatures and Bubble Profiles with Multiple Fields. Comput. Phys. Commun., 183:2006–2013, 2012.DOI
  27. [27]Mariano Quiros. Finite temperature field theory and phase transitions. In ICTP Summer School in High-Energy Physics and Cosmology, pages 187–259, 1999.
  28. [28]Peter Brockway Arnold and Olivier Espinosa. The Effective potential and first order phase transitions: Beyond leading-order. Phys. Rev. D, 47:3546, 1993.
  29. [29]Rajesh R. Parwani. Resummation in a hot scalar field theory. Phys. Rev. D, 45:4695, 1992.
  30. [30]Joan Elias-Miró, Jose R. Espinosa, and Thomas Konstandin. Taming Infrared Divergences in the Effective Potential. JHEP, 08:034, 2014.
  31. [31]Stephen P. Martin. Taming the Goldstone Contributions to the Effective Potential. Phys. Rev. D, 90(1):016013, 2014.DOI
  32. [32]D. A. Knoll and D. E. Keyes. Jacobian-free Newton-Krylov methods: A Survey of approaches and applications. J. Comput. Phys., 193(2):357–397, 2004.DOI
  33. [33]A. H. Baker, E. R. Jessup, and T. A. Manteuffel. A Technique for Accelerating the Convergence of Restarted GMRES. SIAM J. Matrix Anal. Appl., 26(4):962–984, 2005.
  34. [35]Ameya D. Jagtap and George Em Karniadakis. Extended Physics-informed Neural Networks (XPINNs): A Generalized Space-Time Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations. Commun. Comput. Phys., 28(5):2002–2041, 2020.
  35. [36]Xuhui Meng and George Em Karniadakis. A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems. J. Comput. Phys., 401:109020, 2020.DOI
  36. [37]James Kirkpatrick, Brendan McMorrow, David H. P. Turban, Alexander L. Gaunt, James S. Spencer, Alexander G. D. G. Matthews, Annette Obika, Louis Thiry, Meire Fortunato, David Pfau, Lara Román Castellanos, Stig Petersen, Alexander W. R. Nelson, Pushmeet Kohli, Paula Mori-Sánchez, Denis Hassabis, and Aron J. Cohen. Pushing the frontiers of density functionals by solving the fractional electron problem. Science, 374(6573):1385–1389, 2021.DOI
  37. [38]Nasim Rahaman, Aristide Baratin, Devansh Arpit, Felix Draxler, Min Lin, Fred A. Hamprecht, Yoshua Bengio, and Aaron Courville. On the Spectral Bias of Neural Networks. In Proceedings of the 36th International Conference on Machine Learning (ICML), volume 97 of PMLR, pages 5301–5310, 2019.
  38. [39]Zhi-Qin John Xu, Yaoyu Zhang, Tao Luo, Yanyang Xiao, and Zheng Ma. Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks. Commun. Comput. Phys., 28(5):1746–1767, 2020.DOI
  39. [40]Sifan Wang, Xinling Yu, and Paris Perdikaris. When and why PINNs fail to train: A neural tangent kernel perspective. J. Comput. Phys., 449:110768, 2022.
  40. [41]Ben Moseley, Andrew Markham, and Tarje Nissen-Meyer. Finite Basis Physics-Informed Neural Networks (FBPINNs): a scalable domain decomposition approach for solving differential equations. Adv. Comput. Math., 49(4):62, 2023.DOI
  41. [42]Nikhil Vyas, Depen Morwani, Rosie Zhao, Itai Kaplun, Sham Kakade, and Boaz Barak. SOAP: Improving and Stabilizing Shampoo using Adam. arXiv:2409.11321, 2024.arxiv.org/abs/2409.11321
  42. [43]Diederik P. Kingma and Jimmy Ba. Adam: A Method for Stochastic Optimization. arXiv:1412.6980, 2014.arxiv.org/abs/1412.6980
  43. [44]Vineet Gupta, Tomer Koren, and Yoram Singer. Shampoo: Preconditioned Stochastic Tensor Optimization. In Proceedings of the 35th International Conference on Machine Learning, volume 80, pages 1837–1845, 2018.arxiv.org/abs/1802.09568
  44. [45]Ilya Loshchilov and Frank Hutter. SGDR: Stochastic Gradient Descent with Warm Restarts. arXiv:1608.03983, 2016.arxiv.org/abs/1608.03983
  45. [46]Matthew Gonderinger, Yingchuan Li, Hiren Patel, and Michael J. Ramsey-Musolf. Vacuum Stability, Perturbativity, and Scalar Singlet Dark Matter. JHEP, 01:053, 2010.
  46. [47]Marie E. Machacek and Michael T. Vaughn. Two Loop Renormalization Group Equations in a General Quantum Field Theory. 1. Wave Function Renormalization. Nucl. Phys. B, 222:83–103, 1983.
  47. [48]Marie E. Machacek and Michael T. Vaughn. Two Loop Renormalization Group Equations in a General Quantum Field Theory. 2. Yukawa Couplings. Nucl. Phys. B, 236:221–232, 1984.
  48. [49]Marie E. Machacek and Michael T. Vaughn. Two Loop Renormalization Group Equations in a General Quantum Field Theory. 3. Scalar Quartic Couplings. Nucl. Phys. B, 249:70–92, 1985.

Paper details

vlnc:W_u82hwdsn
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 4

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents