REVIEW
Spectral functions at small energies and the electrical conductivity in hot, quenched lattice QCD
High Energy Physics - Lattice · Submitted 12 Mar 2007 · 20 references
Pith's verdict: The kernel-rescaling fix for Bryan's MEM is real, simple, and worth stealing; the σ/T = 0.4 number is a low-ω spectral-weight estimate dressed as a transport coefficient, and should be cited as such.
T0 · AI review
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
The Maximum Entropy Method's small-energy instability at finite temperature comes from a kernel that diverges as $1/ω$, and rescaling the kernel and spectral function by $ω/2T$ removes it, allowing a nonzero zero-energy intercept and an电气 (
The reading
What carries the argument
The load-bearing device is the rescaling of the MEM kernel, Eq. (8): $K(ω,τ) = (ω/2T)K(ω,τ)$ and $ρ(ω) = (2T/ω)ρ(ω)$, so that $Kρ = Kρ$ point by point while the new kernel is regular at the origin. Because the singular vectors of the rescaled kernel are finite at $ω=0$, the exponentiated parametrization $ρ(ω) = m(ω) exp(Σ_i c_i u_i(ω))$ can be evaluated at zero energy, and the intercept — which by the Kubo formula is proportional to the transport coefficient — becomes a fitted quantity rather than an extrapolation. The companion piece is the default model $a^2 m(ω) = m_0(b + aω)$ with $b ≲ 1$, whose role is to allow a nonzero intercept at all; the authors report that taking $b=0$ makes the routine fail to converge, and they check $b = 0.1, 0.5, 1.0$ and $N_ω = 1000, 2000$ in the low-energy region.
Extended reading notes
Core claim
The authors show that the kernel $K(ω,τ) = cosh[ω(τ-1/2T)]/sinh(ω/2T)$ has the small-$ω$ expansion $2T/ω + (ω/T)[1/6 - τ T(1-τ T)] + O(ω^3 T^3)$, so it diverges as $ω→0$ and the singular value decomposition that defines Bryan's subspace produces basis functions $u_i(ω)$ that all diverge at small energy. This is offered as the explanation for a familiar symptom in finite-temperature MEM studies: the value of $ρ$ at the smallest nonzero $ω$ sits inconsistently with larger energies and drifts with the discretization $Δω$, and the routine sometimes fails to converge. Rewriting $K = (ω/2T)K$ and $ρ = (2T/ω)ρ$ leaves $Kρ$ invariant, gives $K(0,τ)=1$, and yields basis functions with a well-defined finite value at $ω=0$; the point $ω=0$ is then included in the fit. With a default model $a^2 m(ω) = m_0(b + aω)$ that permits a finite intercept ($b ≲ 1$), the reconstructed vector spectral function shows, on the $N_τ=24$ lattice at $T/T_c = 1.5$ and the $N_τ=16$ lattice at $T/T_c = 2.25$, a nonzero intercept at $ω=0$ that varies smoothly over $0 ≲ ω/T ≲ 4$, whereas on the cold lattice the intercept is an order of magnitude smaller. Dividing the intercept by six gives $σ/T = 0.4 ± 0.1$, normalized to a single flavour.
Load-bearing premise
The load-bearing premise is that the zero-energy intercept returned by the fit is a property of the correlator data rather than of the default model $m(ω) = m_0(b + aω)$, whose parameter $b$ directly controls whether a nonzero intercept is permitted at all; the paper shows curves for $b = 0.1, 0.5, 1.0$ but does not tabulate the fitted intercept separately for each, so the reader cannot see how much of $σ/T = 0.4$ is prior-driven.
Editorial extensions
If this is right
- The same rescaling can be applied to any channel whose kernel shares the $1/ω$ divergence, so shear and bulk viscosity extractions from lattice correlators inherit a repaired low-energy reconstruction rather than an unstable one.
- Because the fix is a change of variables rather than a change of physics, previously published finite-temperature MEM spectral functions can be re-examined for whether their small-$ω$ behaviour was an artifact of the divergent kernel.
- The reported $σ/T = 0.4 ± 0.1$ per flavour, multiplied by the sum of squared electric charges of the light flavours, gives a physical conductivity that can be compared with transport models of heavy-ion collisions and with weak-coupling estimates.
- Repeating the analysis with Wilson fermions and the conserved point-split current, and including dynamical quarks on anisotropic lattices, would test whether the intercept survives away from the staggered, quenched setting.
Reading between the lines
- The paper's own diagnostic suggests a broader reading: whenever a lattice kernel is singular at a point where the physics is smooth, the instability may be traced to the basis rather than to the data, and a variable redefinition that regularizes the kernel is a cheaper remedy than adding data or tightening priors.
- The reported intercept sits at $ω/T ≲ 4$, well above the weak-coupling scale $ω/T ∼ g^4(T)$ where perturbation theory places the transport peak; if the true transport peak is that narrow, a lattice reconstruction with $O(10)$ time slices may be fitting a broadened remnant rather than the peak itself, which would make the numerical value of $σ/T$ a scheme-dependent quantity.
- A natural cross-check the authors do not perform here is to fit the same correlators with an explicit Breit-Wigner or hard-thermal-loop ansatz and compare the extracted intercept with the MEM result; agreement would strengthen the claim that the intercept is data-driven, disagreement would locate the residual prior dependence.
- The stated error bar is described as an indication of MEM reconstruction uncertainty rather than a statistical one, so a bootstrap or jackknife over gauge configurations would be the obvious next step to separate the two.
Original abstract
In lattice QCD, the Maximum Entropy Method can be used to reconstruct spectral functions from euclidean correlators obtained in numerical simulations. We show that at finite temperature the most commonly used algorithm, employing Bryan's method, is inherently unstable at small energies and give a modification that avoids this. We demonstrate this approach using the vector current-current correlator obtained in quenched QCD at finite temperature. Our first results indicate a small electrical conductivity above the deconfinement transition.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
In lattice QCD, the Maximum Entropy Method can be used to reconstruct spectral functions from euclidean correlators obtained in numerical simulations. We show that at finite temperature the most commonly used algorithm, employing Bryan's method, is inherently unstable at small energies and give a modification that avoids this. We demonstrate this approach using the vector current-current correlator obtained in quenched QCD at finite temperature. Our first results indicate a small electrical conductivity above the deconfinement transition.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | In lattice QCD, the Maximum Entropy Method can be used to reconstruct spectral functions from euclidean correlators obtained in numerical simulations. | background 0.99 | 0.06 | 0.03 | asserted 0.60 |
| s1 | We show that at finite temperature the most commonly used algorithm, employing Bryan's method, is inherently unstable at small energies and give a modification that avoids this. | contribution 0.88 | 0.94 | 0.26 | asserted 1.00 |
| s2 | We demonstrate this approach using the vector current-current correlator obtained in quenched QCD at finite temperature. | method 0.75 | 0.91 | 0.04 | asserted 1.00 |
| s3 | Our first results indicate a small electrical conductivity above the deconfinement transition. | result 1.00 | 0.95 | 0.67 | hedged possibility 0.99 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
Figures
Showing 3 of the paper's 4 figures. See all figures in the PDF ↗ Figures reproduced from arXiv:hep-lat/0703008 by the authors.
References
How later papers used it
72 later papers cite it
1 later paper by other authors is listed below: 1 used its method or tool. 3 more papers by the same authors are listed: 2 built on it and 1 used its method or tool.
Builds on it 2 papers · 2 by the same authors
By the same authors
-
Extended this paper's work further by using twisted boundary conditions with the same choices of twists as another work.
Polarised correlators at finite temperature (2010) (same authors) · 0.95 · paraphrased: the licence does not allow quoting
-
Extended this paper by including non-zero momenta.
Longitudinal and transverse meson correlators in the deconfined phase from the lattice (2010) (same authors) · 1.00 · paraphrased: the licence does not allow quoting
Uses its method or tool 2 of 11 papers with this label · 1 by the same authors
-
Solved its underdefined linear system using this paper's Maximal Entropy Method.
Quark mass dependence of the vacuum electric conductivity induced by the magnetic field in SU(2) lattice gluodynamics (2010) · 1.00 · paraphrased: the licence does not allow quoting
By the same authors
-
Attempted to study this using this paper's default model that explicitly allows the limit of the spectral function over omega to be nonzero.
Charmonium properties in the quark-gluon plasma (2007) (same authors) · 0.97 · paraphrased: the licence does not allow quoting
How the literature uses it
63 arXiv papers cite this in 102 sentences
Papers citing it · 53 sentences without a confident label
Builds on it 2 papers
-
We extend the work in Ref. [ref] further by using twisted boundary conditions with the same choices of twists as in Ref. [ref] .
Polarised correlators at finite temperature · Lattice and Fitting Details · 0.95
-
This work is an extension to our earlier work [ref] to non-zero momentum.
Longitudinal and transverse meson correlators in the deconfined phase from the lattice · Lattice Background · 1.00
Uses its method or tool 9 papers
-
Namely, we extract electric conductivity from the Euclidean correlator of two vector currents $j_ix = q̅x γ_i qx$: [equation] where $ρ_ijw$ is the corresponding spectral function, $Kw, τ = w/2 T coshw τ- 1/2 T/sinhw/2 T$ [ref] , $T = N_t a^-1$ is the temperature, $a$ is the lattice spacing and $N_t$ is the size of the lattice in the time direction.
Quark mass dependence of the vacuum electric conductivity induced by the magnetic field in SU(2) lattice gluodynamics · Introduction · 1.00
-
Spectral functions were then determined using the maximum entropy method (MEM) [ref] , with the modified kernel introduced in [ref] , which greatly improved the stability and convergence of the MEM analysis.
Charmonium properties in the quark-gluon plasma · Methods · 1.00
-
These relations are inverted with the help of the Maximum Entropy Method [ref] .
Numerical study of the conductivity of graphene monolayer within the effective field theory approach · Introduction · 1.00
Uses its data or values 4 papers
-
In Ref. [ref] we determined the electrical conductivity, $σ/T = 0.4 ±0.1$, a result which has since been confirmed [ref] .
Polarised correlators at finite temperature · Lattice and Fitting Details · 0.87
-
We take a constant permittivity $ϵ= 1$, a constant permeability $μ= 1$ and a constant finite electrical conductivity $σ_0 = 5.8 MeV$ [ref] .
Charge-dependent anisotropic flow in high-energy heavy-ion collisions from relativistic resistive magneto-hydrodynamic expansion · Electromagnetic field · 0.97
-
At RHIC $γ=100$, $σ≈5.8$ MeV [ref] .
Time and space dependence of electromagnetic field in relativistic heavy-ion collisions · Introduction · 0.99
Compares with it 6 papers
-
In Fig. [ref] we collect and compare different determinations of $σ_el$ from recent lattice determinations using staggered quenched (grey) [ref] , Wilson-Clover quenched (red) [ref] , Wilson-Clover quenched in the continuum limit (blue triangles and magenta bars) [ref] , $n_f=2+1$ (green whiskerbars) [ref] , as well as our previous $n_f=2$ (blue) Wilson-Clover studies [ref] .
Charge transport and vector meson dissociation across the thermal phase transition in lattice QCD with two light quark flavors · Electrical conductivity across the phase transition · 0.99
-
As $𝒞> N_f logT$ and the charged-density part contribution is suppressed at the leading order behaviour for total DC conductivity ($σ_DC$) in eq([ref]), it is evident that $σ_DC/T≈Constant$, which is consistent with results of the bottom-up holographic QGP from VQCD in [ref] , and lattice QGP plots displayed in the red [ref] and blue [ref] curves in Figure(2) of [ref] .
Charge Transport in Magnetized Holographic $\mathcal{M}$-QGP · Conclusion and Future Directions · 0.96
-
This agrees with [ref] , but is at odds with an earlier calculation [ref] .
Particle production in strong electromagnetic fields in relativistic heavy-ion collisions · Liénard-Wiechert potentials in static medium · 0.98
Cites it as background 12 papers
-
Nonetheless, increasing attention has been given to the problem, resulting in a growing literature of computational strategies [ref] , and applications in different contexts of particle physics, such as the computation of scattering amplitudes [ref] , inclusive decays [ref] , spectroscopy [ref] and QCD at finite temperature [ref] .
Bayesian solution to the inverse problem and its relation to Backus-Gilbert methods · Introduction · 1.00
-
I will not describe the calculations of the electric conductivity [ref] or heavy quark diffusion constant [ref] , except for saying that they share many features with the viscosity studies.
Transport properties of the quark-gluon plasma from lattice QCD · Introduction · 0.99
-
In equilibrium, there have been attempts to extract transport coefficients also using nonperturbative lattice QCD methods [ref] .
Heavy quark diffusion in an overoccupied gluon plasma · Introduction · 0.97
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.