Worm Algorithms for the Schwinger Model

arXiv:2611.05402 · hep-lat · Submitted 2026-10-18 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: High Energy Physics: Lattice. The day in lattice field theory: what the simulations put on the table.

Kai: We're looking at the paper: "Worm Algorithms for the Schwinger Model".

Mira: Here's the gist of the paper. This thesis studies the Schwinger model in (1 + 1) dimensions using a lattice formulation and its dual representation.

Kai: The main objective is to investigate the numerical behaviour of the model in the presence of a fermion mass and a topological angle.

Mira: Model Formulation and Dual Representation The study employs a lattice formulation for the Schwinger model, which involves introducing link variables and staggered fermions to handle the discretization.

Kai: The dual representation replaces original degrees of freedom with configurations of plaquettes, dimers and, in the massive case, monomers.

Title and authors: Kai: So we're looking at this paper, "Worm Algorithms for the Schwinger Model," and it's about using a specific way to simulate a simple but important model in one plus one dimensions.

Mira: It sounds like they are taking the Schwinger model and putting it on a lattice to see how things behave when you add a fermion mass and some kind of topological angle.

Lev: So, what's the core idea here? Is this just another way to simulate quantum field theory on a grid, or is there something fundamentally new about the approach they are taking?

Kai: They use a lattice formulation and then switch to a dual representation. Instead of dealing with the original fields directly, they replace them with configurations made up of plaquettes, dimers, and in the massive case monomers.

Mira: That's interesting because for the massless model, this dual representation helps get around some problems when you try to run Monte Carlo simulations by removing that sign problem entirely.

Lev: Removing the sign problem is a big deal for any simulation trying to get reliable numbers on real hardware because those kinds of updates are often computationally very expensive and prone to getting stuck in bad areas of the configuration space.

Kai: Exactly, and for the massive model, they also analyze how that sign problem shows up when you try to estimate physical quantities like the average plaquette or the chiral condensate through reweighting.

Mira: And they are using Monte Carlo algorithms specifically tailored to this dual representation, including both local and non-local updates, which lets them sample configurations that are allowed by the model's constraints.

Lev: That combination of local and non-local updates is what makes exploring the configuration space feasible for a simulation; it allows you to make changes in the gauge flux and fermion configurations in different ways.

Kai: The paper then explores how this setup behaves when you vary two key parameters: the dimensionless mass divided by the coupling constant, and this topological angle theta.

Title and authors: Mira: The results show that when they look at heat maps of observables like the average plaquette and the chiral condensate, there's a really strong dependence on that topological angle near theta equals pi, specifically where theta over pi is close to one.

Lev: That dependence near pi suggests something non-trivial happening in the vacuum structure when you increase the fermion mass; it implies a rearrangement of what the vacuum looks like in that specific region.

Kai: The paper points out that while the average plaquette shows this strong dependence, the chiral condensate just increases steadily with mass for all angles they tested.

Mira: They also noted that near theta equals pi, the topological charge density shows a much sharper variation as you increase the fermion mass, which hints at some nonanalytic behavior in the limit where you consider infinitely large space.

Lev: That nonanalytic hint is something we need to be careful about when we try to translate this to actual hardware because those kinds of sharp changes can make convergence really slow or cause issues with error estimation.

Kai: They also pointed out a major limitation: the sign problem gets really bad where the average sign is small, which limits how much quantitative data they can get in those specific parameter regions.

Mira: So, while the dual formulation gives us access to non-trivial signatures of this massive theory with that topological term, there's still a hurdle in getting precise numbers because of the sign problem.

Lev: The paper suggests that increasing the fermion mass could lead to an abrupt rearrangement of the vacuum near theta equals pi, but we need better statistical control before we can trust those kinds of conclusions.

Kai: Looking ahead, they suggest future work needs to focus on improving statistical and algorithmic control, optimizing how fast those updates run, and finding ways to deal with that sign problem directly.

Mira: The dual formulation is useful for studying the Schwinger model numerically in the presence of a topological angle because it simplifies things enough to allow these kinds of simulations.

Title and authors: Lev: The idea that this formulation can show a rearrangement near theta equals pi is valuable, but we need to see if that effect actually manifests clearly when you move from this simple model onto more realistic systems.

Kai: It seems like the main contribution here is showing how the dual representation lets you probe those interesting features of the massive theory with a topological term in a way that's otherwise hard.

Mira: We should also keep an eye on how that topological charge density varies near pi as mass increases, because if that nonanalytic behavior holds, it might point to something deeper about how gauge fields interact with matter in these low dimensions.

Lev: From an engineering standpoint, the reliance on these complex update algorithms means that getting a good signal from this simulation requires extremely efficient code and careful monitoring of those statistical errors they estimate using the jackknife method.

Kai: So, to wrap up, the paper "Worm Algorithms for the Schwinger Model" shows that we can use a dual representation with worm algorithms to study how fermion mass and topological angles affect the vacuum structure in this model.

Mira: The key finding is that increasing mass near theta equals pi causes a sharper change in topological charge density, suggesting a possible vacuum rearrangement, but we still have the sign problem limiting quantitative measurements.

Lev: It’s useful because it gives us a roadmap for what kinds of algorithms to build next if we want to simulate these types of theories on actual quantum hardware.

Kai: We should definitely keep an eye on how they tackle the sign problem in the future, because that's where the real difficulty lies when you try to get definitive answers about phase transitions or continuum limits.

Mira: That’s right, and we need to see if that rearrangement near pi is something we can actually observe with better control over those statistical errors.

Lev: We'll keep following this work because the dual formulation opens up new avenues for understanding these types of models, even if the current sign problem limitations make definitive claims tricky right now.

The paper's summary: Kai: So, this paper is basically showing how you can use a dual representation for the Schwinger model on a lattice to handle fermion mass and topological angles better than before.

Mira: Right, so instead of just looking at the original fields directly, they rewrite everything in terms of plaquettes and dimers—and monomers when you have that fermion mass included.

Lev: That's the setup they use to get around some problems with sampling; for example, this dual formulation actually removes the sign problem when there is no fermion mass involved.

Kai: But for their massive version, they still run into issues where the average sign gets small, which limits how much precise data you can get.

Mira: That's true, and that's where they use a diagnostic tool called the average sign to show exactly which parameter regions are hitting those hard corners of the problem.

Lev: From an engineering standpoint, it means that when you try to run this on real hardware, you need very careful monitoring because if the average sign drops too low, your sampling will just fail.

Kai: They map out a grid of mass and angle values using heat maps to show how things like the average plaquette and the chiral condensate depend on those parameters together.

Mira: And what they find is that when you look at that map, the average plaquette changes really strongly when you get close to theta equals pi, specifically where theta over pi is one.

Lev: That strong dependence near pi suggests a possible rearrangement of the vacuum structure when the fermion mass increases in those specific conditions.

Kai: They also mention that while the average plaquette shows this sharp variation, the chiral condensate just climbs steadily with more fermion mass across all angles they tested.

Mira: And they point out that near theta equals pi, the topological charge density gets much sharper as you increase the fermion mass, which hints at some nonanalytic behavior in the infinite space limit.

Lev: That nonanalytic hint is interesting because it suggests something deeper about how those gauge fields interact with matter when you push things to their limits.

Kai: So, while this dual formulation is useful for seeing signatures of a massive theory with a topological term, the sign problem really stops them from getting definitive quantitative numbers in the most interesting regions.

Mira: Exactly, and they conclude that while increasing the fermion mass might cause an abrupt vacuum change near theta equals pi, they still need better statistical control to prove it.

Lev: Future work has to focus on improving that statistical control and finding ways to mitigate the sign problem itself before this kind of insight can be fully trusted.

Kai: So, the dual representation gives us a way to see these complex features, but we're still waiting on better algorithms and fewer sign problem headaches.

The paper's improvements: Kai: So, this paper isn't just showing what they built; they’re suggesting how to make those simulations even better by improving their update algorithms.

Mira: Right, so the authors point out that combining plaquette and worm updates lets them explore the configuration space more effectively than using either one alone.

Lev: That makes sense from an engineering viewpoint because you need both local changes for fine-tuning and non-local changes to move around the configuration space efficiently.

Kai: They explain that this combined approach lets them do both local and non-local updates, which helps them sample configurations that are actually allowed by the model's rules.

Mira: And they also suggest a way to calculate things like the chiral condensate more directly by differentiating the dual partition function with respect to the fermion mass.

Lev: That would mean calculating it as one over V times the derivative of that partition function with respect to mass, related to counting those monomers in their dual picture.

Kai: It’s a nice way to get a better handle on how the condensate changes when you nudge that fermion mass parameter around.

Mira: And they also suggest that monitoring the average sign is a good diagnostic tool because it shows you where the sign problem gets most difficult to handle.

Lev: That helps us pinpoint exactly which part of the simulation is breaking down so we know where to focus our error correction efforts on real hardware.

Kai: They are essentially giving us a better toolkit for running these kinds of simulations, even when they're dealing with those tricky topological angles.

Mira: So, it’s about making the exploration process smarter by using these dual updates and better ways to calculate observables like the condensate.

Lev: It moves the goal from just getting *a* number to getting a more reliable estimate of how that number changes under varying conditions.

Kai: This suggests that if we want to simulate these models on real quantum hardware, we need algorithms that can efficiently navigate this kind of configuration space.

Conclusion: Kai: So, to wrap up, this paper "Worm Algorithms for the Schwinger Model" shows that using a dual representation with these specific update algorithms helps us probe the massive Schwinger model's behavior under different mass and topological angle settings.

Mira: Right, it gives us a new way to see things, especially around that topological angle of pi where the vacuum structure gets interesting when the fermion mass is high.

Lev: From an error correction standpoint, it’s useful because if we can build algorithms that work this well on a computer, those methods are directly applicable to making real quantum computers run more efficiently.

Kai: So what does this mean for us? It suggests that understanding how gauge fields and matter interact in low dimensions with topological terms is something we can get better numbers on.

Mira: And it points toward the possibility of a specific vacuum rearrangement when you increase fermion mass near theta equals pi, which is a physical idea we can actually test on the lattice.

Lev: That physical rearrangement would be hard to confirm if our simulation methods are still too noisy or if the sign problem is too bad to get clean data.

Kai: Exactly, so while this gives us a new window into the massive theory, we still have that sign problem limiting how precise those conclusions can be right now.

Mira: So future work has to focus on fixing those statistical hurdles and maybe finding better ways to handle the sign problem when studying these topological features.

Lev: That’s the next step, and if we can get the overhead down for running these complex updates on real quantum hardware, then this kind of analysis becomes much more practical.

Kai: Yeah, so it's a solid piece of work showing how to map out those tricky regions in the parameter space for the massive Schwinger model.

Mira: It really highlights that the dual formulation is a powerful tool when you need to see signatures that you just can't see with the original fields.

Lev: We should keep an eye on how they tackle those statistical errors in their next work because getting reliable results is everything for us building error correcting codes.

Lucas Ramíz Zalduendo, Alejandro Vaquero Avilías-Casco

hep-lat

Submitted: 2026-10-18

Updated: 2026-10-18

Comments: 44 pages, 16 figures. Bachelor's thesis (BSc in Physics), University of Zaragoza, academic year 2025-2026. Supervisor: Alejandro Vaquero Avilés-Casco

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 77/100

The gist: This thesis studies the Schwinger model in (1 + 1) dimensions using a lattice formulation and its dual representation.

Key concepts

Dual Representation
This method replaces the original gauge field degrees of freedom with configurations like plaquettes and dimers. For massless models, this simplifies sampling by removing the sign problem and allows for easier exploration of configuration space through updates.
Worm Algorithms
These are specific Monte Carlo update moves designed to efficiently sample admissible configurations in the dual representation. Combining plaquette and worm updates enables both local and non-local changes, which is crucial for effectively exploring the complex configuration space of the model.
Topological Angle ($ heta$)
The topological angle $\theta$ introduces a background field that influences the system's behavior. In this context, it acts like an electric field that can be screened by fermion-pair production, leading to distinct physical effects depending on its value.
Average Sign ($\langle sign \rangle$)
The average sign is a diagnostic observable used to monitor the difficulty of Monte Carlo simulations. Regions where the average sign is small indicate areas where the sign problem is most severe, limiting reliable quantitative measurements in those specific parameter regimes.

Terminology

Summary

This thesis studies the Schwinger model in (1 + 1) dimensions using a lattice formulation and its dual representation. p. 1 The main objective is to investigate the numerical behaviour of the model in the presence of a fermion mass and a topological angle. p. 1

Model Formulation and Dual Representation

The study employs a lattice formulation for the Schwinger model, which involves introducing link variables and staggered fermions to handle the discretization. The dual representation replaces original degrees of freedom with configurations of plaquettes, dimers and, in the massive case, monomers. p. 1 For the massless model, this dual formulation removes the sign problem and provides a configuration space for update algorithms. p. 1 The partition function in the dual representation is given by a sum over loops, dimers and plaquette occupations subject to constraints on Grassmann variables and net gauge flux. p. 5

Simulation Methodology

Monte Carlo algorithms are implemented adapted to the dual representation, including local and non-local updates that allow admissible configurations to be sampled. Physical quantities such as the average plaquette, dimer density, chiral condensate and the average sign are computed from these simulations. p. 1 Thermalisation is monitored through observables like the average plaquette ⟨Up⟩ and dimer density ⟨Nd/V ⟩. p. 1 Error estimation is performed using the jackknife method to account for autocorrelation. p. 2

Exploration of Parameter Space

Simulations are conducted on a grid of values for the dimensionless mass m/g and topological angle θ. Heat maps are used to examine the joint dependence of observables like the average plaquette ⟨Up⟩ and chiral condensate ⟨χχ¯ ⟩ on these parameters. p. 3 The heat maps reveal that the average plaquette has a more pronounced dependence on the topological angle, with higher values near θ/π ≃ 1. p. 3 The chiral condensate increases monotonically with mass for all angles considered. p. 1

Physical Interpretation and Limitations

The results suggest a possible rearrangement of the vacuum around θ = π when the fermion mass is increased, which can be interpreted qualitatively in terms of the Schwinger effect. The topological charge density ⟨q⟩ shows a sharper variation near θ = π as mass increases, suggesting nonanalytic behaviour in the thermodynamic limit. p. 3 However, the sign problem becomes most significant where the average sign is small, limiting quantitative measurements in this region. p. 1 The degradation of the average sign prevents reliable quantitative conclusions about a transition or its continuum extrapolation. p. 2

Conclusions and Outlook

The dual formulation is useful for studying the Schwinger model numerically in the presence of a topological angle. The study suggests that increasing the fermion mass may lead to an abrupt rearrangement of the vacuum near θ = π. p. 5 Future work should focus on improving statistical and algorithmic control, optimizing update efficiency, and investigating methods for mitigating the sign problem. p. 1 The dual formulation gives access to non-trivial signatures of the massive theory with a topological term and suggests a possible rearrangement of the vacuum around θ = π. p. 1

Appendix A: Bosonisation

Bosonisation provides an alternative way to study the Schwinger model by rewriting the fermionic theory in terms of a real scalar field. The massless bosonised Lagrangian is L = 1/2∂µϕ ∂µϕ + 1/2E squared + g√πϕE + gθ2/2πE. p. 2 This formulation shows that the electric field responds to two contributions, one from the bosonic field ϕ and one from the topological angle θ. p. 4 The massive model introduces a competition between the quadratic term from gauge fields and a periodic term from the fermion mass, which can lead to multiple competing minima near θ = π.

Appendix B: Discretisation

The lattice action is constructed by discretising the gauge, fermionic and topological terms using link variables and staggered fermions. The gauge contribution is expressed as SG = βXn[1 - ReUp(n)], with β = 1/g2a2, which recovers the continuum action in the limit a → 0. p. 9 The topological term is defined as Sθ = iθQlat[U] = θ/4πXn(Up(n) - U−1p(n)). p. 2

Appendix C: Update Algorithm Pseudocode

The simulation employs a combination of plaquette and worm updates to explore admissible configurations. The closed worm update involves propagating endpoints until they meet again, while the open worm update allows for creation or annihilation of monomers with probabilities p+ and p−. p. 2

Appendix D: Auxiliary Figures

The auxiliary figures illustrate thermalisation in the (m/g, θ) plane and slices of the parameter space. Figure 9 shows heat maps of the topological charge density ⟨q⟩ and topological susceptibility χq in the (θ/π, m/g) plane. p. 3 These figures show that the variation of ⟨q⟩ becomes sharper near θ = π as mass increases. p. 1 Figure 13 shows heat maps of the average sign ⟨sign⟩ in the (θ/π, m/g) plane for two sets of lattice parameters. p. 3 The region with smaller average sign lies near θ/π ≃ 1, where topological observables vary most strongly. p. 1

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved simulation algorithms can efficiently explore admissible configuration space by combining plaquette and worm updates to facilitate both local and non-local changes in gauge flux and fermionic configurations. The text states that Combining plaquette and worm updates allows both local and non-local changes, facilitating exploration of admissible configurations.

  2. The system can now accurately determine the chiral condensate as a function of fermion mass by differentiating the dual partition function with respect to mass, relating it to monomer counting: The chiral condensate and its susceptibility can therefore be defined as ⟨χχ¯⟩ = 1/V ∂/∂m log Z, χχχ¯ = 1/V ∂2/∂m2 log Z.

  3. The AI system can qualitatively interpret the vacuum rearrangement near theta equals pi in the massive Schwinger model by connecting it to physical phenomena: The angle θ acts as a background electric field that can be screened through fermion-pair production.

  4. The system can diagnose regions of severe sign problem difficulty by monitoring the average sign, which is a diagnostic observable: The average sign ⟨sign⟩ can serve as a diagnostic observable, identifying the regions most affected by the sign problem.

Sources

Related papers