A Gaussian Process framework for constraining the nuclear equation of state from microscopic calculations with correlated uncertainties

C. Drischler; J. Kim; T. Zhao; Y. G. Lee

REVIEW 2 major objections 4 minor 150 references

A Gaussian Process framework for constraining the nuclear equation of state from microscopic calculations with correlated uncertainties

T0 review · 2 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that a hierarchical Gaussian process with a two-layer covariance structure can quantify and propagate correlated uncertainties from noisy microscopic many-body calculations to derived nuclear EOS observables such as the…

desk verdict A useful two-level GP framework with honest caveats, but the headline uncertainty bands are unverified under the exchangeability assumption. read the letter →

arxiv 2608.09678 v1 pith:ONQZRWZD submitted 2026-08-10 nucl-th astro-ph.SRnucl-ex

Y. G. Lee , J. Kim , T. Zhao , C. Drischler This is my paper · ORCID

classification nucl-thastro-ph.SRnucl-ex

keywords Gaussianprocessregressionnuclearequationofstatechiraleffectivefieldtheoryuncertaintyquantificationmany-bodyperturbationsymmetryenergycrust-coretransition

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that a carefully structured Gaussian process can turn the spread among six chiral-interaction many-body calculations into statistically useful uncertainty bands for the nuclear equation of state, rather than treating that spread as ordinary noise. The model gives every calculation a share of a common mean EOS and its own correlated deviation from that mean, plus numerical noise; after calibration, the GP predicts a new noise-free EOS and its derivatives with a covariance that includes both the uncertainty in the mean and the interaction-to-interaction scatter. Applied to high-order many-body perturbation theory up to about twice saturation density, the framework constrains the saturation point, the incompressibility and its isospin dependence, the symmetry energy and its slope, and the neutron-star crust-core transition density, with $K_\tau = -423^{+263}_{-374}\,\mathrm{MeV}$ and $n_{\mathrm{cc}} = 0.076^{+0.036}_{-0.023}\,\mathrm{fm}^{-3}$ at 95% credibility. A sympathetic reader would care because these are microscopic, chiral-EFT-based constraints that carry explicit correlated uncertainties across all derived observables, and the same workflow is intended to be reusable on other many-body calculations at zero or finite temperature.

What carries the argument

The central object is a two-layer hierarchical Gaussian process: the latent common mean $\eta \sim \mathcal{GP}(\mu_\eta, k_\eta)$ carries the systematic EOS behavior, and each interaction's deviation $\delta^{(h)}$ is an independent draw from a zero-mean GP with kernel $k_\delta$ (the kernel being the function that sets the prior covariance between any two density–asymmetry inputs). The deviation kernel is $k_\delta = \alpha_{\mathrm{emp}} k_\delta^{(\mathrm{emp})} + k_\delta^{(\mathrm{sm})}$, where the empirical term uses the dominant eigenmodes of the sample covariance and the smooth term is an RBF residual; the load-bearing identity for predictions is $K_{f,*|t} = K_{\eta,*|t} + K_{\delta,*}$, which ensures the uncertainty bands include both mean and interaction-level scatter. A change-surface kernel, an input-dependent blend of several RBF kernels through smooth Gaussian weight functions, is explored for the mean EOS and gives results consistent with a single RBF. GPDiff differentiates the kernel automatically, so the same trained GP jointly samples the energy and arbitrary mixed partial derivatives, feeding derived quantities such as pressure, sound speed, and the Hessian-determinant instability criterion that locates the crust-core transition.

What would settle it

Retrain the GP on five of the six interactions and test whether the held-out sixth interaction's energy-per-nucleon curve is covered by the 68% and 95% predictive bands at the expected rate across all six leave-one-out folds; systematic undercoverage would falsify the exchangeability assumption and the deviation-kernel construction behind the reported credible intervals.

Watch

Extended reading notes

Core claim

The central claim is that the six EOS calculations can be modeled as noisy realizations of a shared latent mean EOS, $\eta(x)$, plus interaction-dependent deviations $\delta^{(h)}(x)$ that are independent draws from a zero-mean GP, plus heteroscedastic white noise. Under this model the predictive distribution for a new interaction's noise-free EOS has mean equal to the posterior mean of $\eta$ and covariance $K_{\eta} + K_{\delta}$, so the final uncertainty bands are wider than the empirical scatter about the sample mean because they also include the uncertainty in the mean itself. The deviation kernel is constructed from the three dominant eigenmodes of the six-interaction sample covariance (interpolated as GPs) plus an RBF residual kernel, and the mean kernel is calibrated to the ensemble average with the reduced deviation covariance $K_{\delta}/H$; with either a single RBF or a change-surface kernel the resulting joint posterior samples give the low-density EOS parameters, neutron-star matter properties, and crust-core transition density reported in Table I, and the GP results agree within their uncertainties with two independent GP-free extraction methods.

Load-bearing premise

The load-bearing premise is that the six chiral interactions are exchangeable draws from a single population of equations of state, so that the spread among them can be summarized by one deviation kernel estimated from their covariance; if the six interactions instead differ systematically (for example, ordered by their resolution scale), the reported credible intervals will be miscalibrated.

Editorial extensions

If this is right

The reported credible intervals for $n_0$, $E_0/A$, $K$, $S_v$, $L$, $K_\tau$, and the crust-core transition properties are joint: all parameters come from the same GP posterior samples, so correlations such as the strong anti-correlation between $n_0$ and $E_0/A$ are part of the output.
Because the predictive covariance includes $K_\delta$, the GP uncertainty bands are wider than the $\pm 2\sigma$ scatter of the six individual calculations; this widening is by construction and the bands match the empirical data when checked against the two GP-free extraction methods.
The kernel choice matters little at these low densities: a single RBF and a three-RBF change-surface kernel give statistically consistent constraints, indicating that a smooth stationary kernel is adequate below about $2n_0$.
GPDiff's arbitrary-order derivative predictions with correlated uncertainties make the same workflow applicable to finite-temperature EOS calculations and to other many-body frameworks, not just the asymmetric-matter MBPT data analyzed here.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the exchangeability assumption in Eq. (24) holds, the same two-stage calibration could be applied to any ensemble of many-body EOS calculations, e.g., different chiral orders or different many-body methods, to separate family-wide spread from method-specific noise; the risk is that six interactions are a very small sample from a population whose spread may not be representative.
The paper does not report a leave-one-interaction-out coverage test; such a test would train on five interactions and check whether the held-out sixth EOS falls inside the 68% and 95% predictive bands at the expected rate, giving a direct check on whether the credible intervals are calibrated.
The inferred $n_{\mathrm{cc}}=0.076^{+0.036}_{-0.023}\,\mathrm{fm}^{-3}$ is consistent with existing chiral-EFT truncation-error estimates, but the authors have not added an explicit truncation-error layer, so the full theoretical uncertainty at densities above $1.5n_0$ is likely larger than the bands shown.
A future measurement of $K_\tau$ from the isoscalar giant monopole resonance in $^{132}\mathrm{Sn}$ provides a direct external check: if the measured value falls far outside $-423^{+263}_{-374}\,\mathrm{MeV}$, the interaction ensemble or the exchangeability assumption would need to be revisited.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear colleague,

Here's my read on arXiv:2608.09678.

The genuinely new piece is the two-level GP model: a common-mean GP for the EOS shared across interactions, an interaction-deviation GP built from the low-rank empirical eigenmodes of the sample covariance, and a white-noise term for MC errors. Wrapped in a JAX package with automatic differentiation for arbitrary derivative predictions, that is a useful and clean formulation. The application to recent asymmetric-matter MBPT data gives the first correlated-uncertainty constraints on Kτ and the crust-core transition density directly from explicit MBPT. The derivations are careful, and the cross-checks against a parametric meta-model and bivariate splines are good practice. I also appreciate how explicitly they state the limitations: no EFT truncation errors, MAP instead of marginalizing hyperparameters, and the i.i.d. assumption.

The main soft spot is exactly what the stress-test note highlights. The predictive covariance in Eq. (52c) treats the six Hebeler interactions as exchangeable draws from a common population. They are not: different SRG scales, 3N cutoffs, and one interaction has altered πN coupling. With H=6, the sample covariance is rank-5 and noisy; each entry has 40-60% relative error. Using three empirical eigenmodes plus an RBF residual estimated from that is a pragmatic choice, but it is not going to give well-calibrated uncertainty bands. The paper does not run any out-of-sample check, like leave-one-interaction-out prediction, to test calibration. So the 95% intervals for Kτ and ncc should be read as conditionally reasonable, not validated. This is a moderate concern, not a fatal one, because the paper frames these as constraints under stated model assumptions and explicitly invites future work. Still, the central claim that it quantifies and propagates correlated uncertainties is stronger than what the evidence supports.

Another softer spot: the 'predictions' are in-sample interpolations, and the reported intervals are lower bounds on theoretical uncertainty because truncation errors are omitted. They say this, so it's not hidden. Just don't quote Table I without that caveat.

Who benefits: anyone modeling the nuclear EOS from a small ensemble of many-body calculations, and anyone building GP tools for derivatives. The GPDiff package could become a real community resource if the code and data ship as promised.

My recommendation: send it to peer review. The framework is novel enough and the physics relevant enough, and the flaws are fixable with a calibration study. A serious referee should push for leave-one-out predictive checks and a discussion of what exchangeability means for a set of chiral interactions.

Referee Report

2 major / 4 minor

Summary. The paper introduces GPDiff, a JAX-based Gaussian process (GP) regression package with automatic differentiation, and applies it to recent asymmetric nuclear matter many-body perturbation theory (MBPT) calculations for six chiral interactions. The statistical model (Sec. II C) decomposes each calculated EOS into a common mean GP η(x), an interaction-dependent deviation GP δ^(h)(x), and heteroscedastic numerical noise. The deviation covariance is calibrated from the sample covariance of the six EOSs via a low-rank empirical kernel plus an RBF residual kernel, and predictions for a new EOS are obtained by combining the posterior of η with the full deviation covariance (Eq. 52). The authors report 95% credible intervals for the saturation point, K, Kτ, Sv, L, the neutron-star EOS, and the crust-core transition density, and they cross-check the results with a parametric meta-model and bivariate splines.

Significance. If the reported credible intervals are well calibrated, the paper provides a useful and general tool for propagating correlated uncertainties from microscopic EOS calculations to derived observables and a set of statistically consistent constraints on low-density nuclear matter. The hierarchical model is clearly derived, and the comparison with two independent extraction methods is a genuine strength. The paper is also candid about limitations, noting in Secs. III B and IV that EFT truncation errors are omitted and that hyperparameters are treated at their MAP values. The central methodological contribution—a reusable, auto-differentiable GP framework with derivative predictions—is potentially of broad utility to the nuclear EOS community. However, the calibrated nature of the uncertainty bands is not yet demonstrated, and the main quantitative claims in Table I rest on an exchangeability assumption that requires additional validation.

major comments (2)

[Sec. II C, Eq. (24) and Eq. (52c)] The assumption that the six interaction-specific deviations δ^(h) are i.i.d. draws from a single zero-mean GP is load-bearing for the predictive covariance K_{f,∗∗|t}, yet the six Hebeler interactions are not an exchangeable random sample: they differ systematically in λ_SRG, Λ_3N, and one of them uses different πN couplings. If, for example, the EOS deviation varies monotonically with λ_SRG, the model will absorb that trend into kδ and report it as variance about a common mean, which would bias the widths of the credible intervals in Table I. The in-sample agreement with the meta-model and spline methods shown in Figs. 4–8 does not test coverage. I recommend adding a leave-one-interaction-out predictive check (train on five interactions, predict the sixth) or an explicit sensitivity analysis of the exchangeability assumption; without such a check, the central claim that Eq. (52c) provides calibrated uncertainty is unverified.
[Sec. II C 3, Eqs. (41)–(45)] The deviation kernel kδ is estimated from a rank-5 sample covariance matrix obtained from H=6 interactions, using only M=3 empirical eigenmodes plus an RBF residual kernel whose hyperparameters are fixed at the restricted-likelihood maximum. The uncertainty in these estimates—particularly in the eigenvalues/eigenvectors of Σδ and in θ_sm—is not propagated into Eq. (52c). With only five residual degrees of freedom, this uncertainty is non-negligible, and because Kδ dominates the predictive covariance, the Table I intervals are likely understated. At minimum, the authors should report how the results change with M (e.g., M=2 vs M=4) and with reasonable perturbations of the RBF hyperparameters, or marginalize over θδ.

minor comments (4)

[Abstract and Table I] The uncertainty bands are presented as constraints without noting that they exclude EFT truncation errors; the caveat appears only in Secs. III B and IV. A one-sentence qualifier in the abstract would prevent misreading.
[Fig. 2 caption and Sec. III] Several places use '2σ confidence level' and 'confidence regions' for Bayesian posterior intervals; 'credibility' would be more appropriate and consistent with the text's own use of 'credibility intervals'.
[Table I] The row for pcc is labeled 'CTT pressure'; this appears to be a typo for 'CCT pressure'.
[Sec. III A, Eqs. (61)–(62)] The covariance matrices in Eqs. (61) and (62) do not explicitly state units; adding units for the entries would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the reported EOS constraints are honest posterior inferences from the same MBPT data used for training, with no fitted parameter masquerading as a prediction.

full rationale

The paper's derivation chain is self-contained and transparent. The training set is the MBPT E/A data of Ref. [16]; the hierarchical model (22) splits it into a common mean η, interaction deviations δ^(h), and numerical noise ε^(h). The deviation kernel kδ is calibrated to the empirical ensemble covariance Σδ via the restricted likelihood (45), and the predictive distribution for a new EOS is assembled in Eq. (52c) as Kf,∗∗|t = Kη,∗∗|t + Kδ,∗∗. This is a data-driven posterior predictive distribution, not a hidden identity: the reported parameters (n0, E0/A, K, Sv, L, Kτ, ncc) are nonlinear functionals of the GP posterior (e.g., Eqs. (54), (57), (60), (68)) and no fitted hyperparameter equals any reported constraint by construction. External empirical ISGMR values and prior EOS ranges are used only for comparison. The few citations to work by the same authors (e.g., Ref. [56] for change-point kernels, Ref. [16] for the MBPT data) are either the data source or a kernel ansatz whose influence the paper checks against a single-RBF kernel (Appendix C), so they do no circular argumentative work. The exchangeability assumption in Eq. (24) and the rank-5 estimate of Σδ are statistical assumptions that affect calibration, but questioning them is a correctness risk, not a circularity.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central inference rests on the exchangeability of six interaction-specific deviations, Gaussian assumptions for deviations and noise, and MC covariance estimates from the same data. No new physical entities are postulated; all introduced objects are statistical (GPs, sample covariance eigenmodes, a new-interaction predictive process).

free parameters (5)

θ_η: mean EOS GP hyperparameters (signal variance, per-dimension length scales) = MAP values not reported in text
Calibrated by maximizing Eq. (50); controls smoothness and amplitude of common mean EOS η, directly shaping derivative predictions and all reported EOS parameters.
θ_δ = {α_emp, θ_sm}: residual deviation kernel hyperparameters = not reported
Estimated by maximizing restricted log-likelihood Eq. (45); sets the RBF regularization and empirical-mode weight in kδ, hence the size of predictive bands.
Empirical eigenmodes λ_m, φ_m of Σδ (m = 1..3) = top 3 eigenvalues of the rank-5 sample covariance
Low-rank empirical kernel Eq. (44) built from MBPT residuals; these are fitted to the same data used for the reported constraints.
Change-surface kernel configuration: centers c_i and smearing B = B = diag(0.5, 0.5), centers at δ_iso = 0, ±1, n = 0.16 fm^-3 (Sec. III)
Chosen fixed values, not calibrated; Appendix C shows results are robust to a single RBF kernel and a 6-kernel CS configuration.
Number of retained empirical modes M = 3 for EOS parameters, 7 for c_s^2 propagation
Model choice; M ≤ H-1 with H=6 interactions; affects how much covariance is represented by the empirical kernel.

assumptions (6)

domain assumption The six interactions are exchangeable draws from a common deviation GP, δ^(h) i.i.d. ~ GP(0,kδ)
Sec. II C, Eq. (24). Load-bearing: defines the uncertainty model for a new EOS; with H=6 it cannot be tested.
domain assumption MBPT energies from Ref. [16] are unbiased realizations of a shared mean EOS plus deviations and diagonal numerical noise
Eqs. (22) and (25); EFT truncation errors are not included, as stated in Sec. III B.
domain assumption EOS is approximately symmetric under δ_iso → −δ_iso; data are reflected about δ_iso=0
Sec. II B; used to double the training set and avoid GP boundary effects at SNM.
standard math Mean function and kernel are sufficiently differentiable for required derivative orders
Sec. II A 3; RBF and CS kernels used are infinitely differentiable, unlike Matérn with finite ν.
ad hoc to paper Zero-mean GP prior for the common EOS is adequate over the training domain
Sec. III states 'we use the zero-mean function'; predictions away from data would revert to zero, but evaluation range is interpolation-dominated.
domain assumption β-equilibrium and Hessian-determinant criterion determine the crust-core transition
Secs. III B and III C, Eq. (68); electron screening and finite-size dynamical corrections are neglected, with expected ~10% shifts noted in text.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A Gaussian Process framework for constraining the nuclear equation of state from microscopic calculations with correlated uncertainties." pith.science (2026). https://pith.science/paper/ONQZRWZD

@misc{pith2026260809678,
  author       = {Pith},
  title        = {Pith review of: A Gaussian Process framework for constraining the nuclear equation of state from microscopic calculations with correlated uncertainties},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ONQZRWZD}},
  note         = {Machine review of arXiv:2608.09678}
}

read the original abstract

We present constraints on the nuclear equation of state (EOS) from microscopic asymmetric matter calculations at zero temperature based on chiral nucleon-nucleon and three-nucleon interactions. The constraints include the saturation point, the isospin dependence of the incompressibility, and the symmetry energy, as well as the crust-core transition density of neutron-star matter. To quantify and propagate correlated uncertainties from noisy many-body calculations to derived observables, we introduce GPDiff, an efficient JAX-based Python package for multivariate Gaussian process (GP) regression with automatic differentiation. After training, GPDiff enables joint predictions of the EOS and derivatives of arbitrary order with respect to the input variables, including mixed partial derivatives. In this initial application, we analyze recent high-order many-body perturbation theory calculations of asymmetric matter up to about twice saturation density and explore nonstationary change-surface kernels, a class of input-dependent kernels, for modeling the EOS. GPDiff is broadly applicable to microscopic nuclear EOS calculations at zero and finite temperature and provides a versatile package for GP-based uncertainty quantification and inference of the nuclear EOS.

Figures

Figures reproduced from arXiv: 2608.09678 by the authors.

**Figure 2.** FIG. 2. Explicit asymmetric matter calculations used for training the GPs. In Ref. [ [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** FIG. 3. Comparison of the one-dimensional GP regression based on different model assumptions (for illustration). Panels (a) [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗

Figures from the paper (10 more)

**Figure 4.** Figure 4: FIG. 4. Posterior distributions of the inferred low-density EOS parameters. The parameters include the saturation density [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]

**Figure 5.** Figure 5: FIG. 5. The predicted saturation density, [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]

**Figure 6.** Figure 6: FIG. 6. Constraints on charge-neutral, [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]

**Figure 7.** Figure 7: FIG. 7. Composition of NSM and instability boundary as [PITH_FULL_IMAGE:figures/full_fig_p018_7.png]

**Figure 8.** Figure 8: FIG. 8. Posterior distributions of the crust-core transition properties and low-density EOS parameters. The crust-core transition [PITH_FULL_IMAGE:figures/full_fig_p019_8.png]

**Figure 9.** Figure 9: FIG. 9. Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p024_9.png]

**Figure 10.** Figure 10: FIG. 10. Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p025_10.png]

**Figure 11.** Figure 11: FIG. 11. Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p026_11.png]

**Figure 12.** Figure 12: FIG. 12. Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p027_12.png]

**Figure 13.** Figure 13: FIG. 13. Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p028_13.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

150 extracted references · 22 canonical work pages

[1]

Ekstr¨ om, G

A. Ekstr¨ om, G. R. Jansen, K. A. Wendt, G. Hagen, T. Papenbrock, B. D. Carlsson, C. Forss´ en, M. Hjorth- Jensen, P. Navr´ atil, and W. Nazarewicz, Phys. Rev. C 91, 051301(R) (2015), arXiv:1502.04682

arXiv 2015
[2]

Ekstr¨ om, G

A. Ekstr¨ om, G. Hagen, T. D. Morris, T. Papenbrock, and P. D. Schwartz, Phys. Rev. C 97, 024332 (2018), arXiv:1707.09028

arXiv 2018
[3]

Drischler, K

C. Drischler, K. Hebeler, and A. Schwenk, Phys. Rev. Lett. 122, 042501 (2019), arXiv:1710.08220

arXiv 2019
[4]

We can therefore determine the common mean process by conditioning on the observed sample mean

Posterior of the common mean EOS With the deviation kernel now calibrated, all compo- nents of the ensemble mean covariance are specified. We can therefore determine the common mean process by conditioning on the observed sample mean. At arbitrary evaluation points, η | ¯yt, xt, x∗ ∼ N mη,∗|t, Kη,∗∗|t , (46) with posterior mean and posterior covariance ma...
[5]

Meta-model

Predictive distribution for a new EOS We now return to the predictive target in Eq. (26). Before conditioning on the MBPT data for the EOS, f (new)(x) is marginally distributed as f (new)(x) ∼ GP(µη(x), kη(x, x′) + kδ(x, x′)) , (51) corresponding to Eq. (31) when also the numerical noise is marginalized over and with the kernels defined in Eqs.(40) and (4...

arXiv
[6]

W. G. Jiang, C. Forss´ en, T. Dj¨ arv, and G. Hagen, Phys. Rev. C 109, 064314 (2024), arXiv:2212.13216

arXiv 2024
[7]

W. G. Jiang, C. Forss´ en, T. Dj¨ arv, and G. Hagen, Phys. Rev. C 109, L061302 (2024), arXiv:2212.13203 [nucl-th]. 24 n0 = 0.170+0.062 −0.038 −21 −18 −15 −12 −9 E0/A E0/A =−15.57+3.04 −4.70 150 200 250 300 350 K K = 223+116 −68 28 32 36 40 Sv Sv = 32.8+7.9 −5.1 40 50 60 70 L L = 51.0+18.7 −11.4 0.120.150.180.210.24 n0 −750−600−450−300−150 Kτ −21−18−15−12−...

arXiv 2024
[8]

Drischler, P

C. Drischler, P. G. Giuliani, S. Bezoui, J. Piekarewicz, and F. Viens, Phys. Rev. C 110, 044320 (2024), arXiv:2405.02748

arXiv 2024

Show all 150 references

[9]

Drischler, J

C. Drischler, J. W. Holt, and C. Wellenhofer, Annu. Rev. Nucl. Part. Sci. 71, 403 (2021), arXiv:2101.01709

2021 arXiv
[10]

Sorensen et al

A. Sorensen et al. , Prog. Part. Nucl. Phys. 134, 104080 (2024), arXiv:2301.13253

2024 arXiv
[11]

Kumar et al

R. Kumar et al. (MUSES), Living Rev. Rel. 27, 3 (2024), arXiv:2303.17021 [nucl-th]

2024 arXiv
[12]

Chatziioannou, H

K. Chatziioannou, H. T. Cromartie, S. Gandolfi, I. Tews, D. Radice, A. W. Steiner, and A. L. Watts, Rev. Mod. Phys. 97, 045007 (2025), arXiv:2407.11153

2025
[13]

Agarwal, J

K. Agarwal, J. Jahan, B. Kardan, P. T. H. Pang, T. Re- ichert, and A. C. Semposki, (2025), arXiv:2511.20378

2025 arXiv
[14]

Epelbaum, H.-W

E. Epelbaum, H.-W. Hammer, and U.-G. Meißner, Rev. Mod. Phys. 81, 1773 (2009), arXiv:0811.1338

2009 arXiv
[15]

Machleidt and D

R. Machleidt and D. R. Entem, Phys. Rept. 503, 1 (2011), arXiv:1105.2919

2011 arXiv
[16]

Hammer, S

H.-W. Hammer, S. K¨ onig, and U. van Kolck, Rev. Mod. Phys. 92, 025004 (2020), arXiv:1906.12122

2020 arXiv
[17]

Epelbaum, H

E. Epelbaum, H. Krebs, and P. Reinert, Front. Phys. 8, 98 (2020), arXiv:1911.11875

2020 arXiv
[18]

Drischler, K

C. Drischler, K. S. McElvain, and P. Arthuis, (2026), arXiv:2603.24532

2026
[19]

Lonardoni, I

D. Lonardoni, I. Tews, S. Gandolfi, and J. Carlson, Phys. Rev. Res 2, 022033(R) (2020), arXiv:1912.09411

2020 arXiv
[20]

Arthuis, C

P. Arthuis, C. Barbieri, F. Pederiva, and A. Roggero, Phys. Rev. C 107, 044303 (2023), arXiv:2203.16167

2023 arXiv
[21]

B. Fore, J. Kim, M. Hjorth-Jensen, and A. Lovato, Com- mun. Phys. 8, 108 (2025), arXiv:2407.21207 [nucl-th]

2025 arXiv
[22]

Lovato, G

A. Lovato, G. Carleo, B. Fore, M. Hjorth-Jensen, J. Kim, A. Rios, and N. Rocco, (2026), arXiv:2602.13826 [nucl- th]

2026
[23]

Duguet, A

T. Duguet, A. Ekstr¨ om, R. J. Furnstahl, S. K¨ onig, and D. Lee, Rev. Mod. Phys. 96, 031002 (2024), arXiv:2310.19419 [nucl-th]

2024 arXiv
[24]

J. A. Melendez, C. Drischler, R. J. Furnstahl, A. J. Garcia, and X. Zhang, J. Phys. G 49, 102001 (2022), arXiv:2203.05528 [nucl-th]

2022 arXiv
[25]

Drischler, J

C. Drischler, J. A. Melendez, R. J. Furnstahl, A. J. Garcia, and X. Zhang, Front. Phys. 10, 92931 (2023), supplemental, interactive Python code can be found on the companion website https://github.com/buqeye/ frontiers-emulator-review , arXiv:2212.04912

2023 arXiv
[26]

P. Cook, D. Jammooa, M. Hjorth-Jensen, D. D. Lee, and D. Lee, Nature Commun. 16, 5929 (2025), arXiv:2401.11694 [cs.LG]

2025 arXiv
[27]

Frame, R

D. Frame, R. He, I. Ipsen, D. Lee, D. Lee, and E. Rrapaj, Phys. Rev. Lett. 121, 032501 (2018), arXiv:1711.07090

2018 arXiv
[28]

Hebeler, J

K. Hebeler, J. M. Lattimer, C. J. Pethick, and A. Schwenk, Astrophys. J. 773, 11 (2013), arXiv:1303.4662 [astro-ph.SR]

2013 arXiv
[29]

J. E. Lynn, I. Tews, J. Carlson, S. Gandolfi, A. Gezerlis, K. E. Schmidt, and A. Schwenk, Phys. Rev. Lett. 116, 062501 (2016), arXiv:1509.03470 [nucl-th]

2016 arXiv
[30]

J. W. Holt and N. Kaiser, Phys. Rev. C 95, 034326 (2017), arXiv:1612.04309

2017 arXiv
[31]

Drischler, S

C. Drischler, S. Han, J. M. Lattimer, M. Prakash, S. Reddy, and T. Zhao, Phys. Rev. C103, 045808 (2021), arXiv:2009.06441 [nucl-th]

2021 arXiv
[32]

S. Huth, C. Wellenhofer, and A. Schwenk, Phys. Rev. C 103, 025803 (2021), arXiv:2009.08885 [nucl-th]

2021 arXiv
[33]

Drischler, S

C. Drischler, S. Han, and S. Reddy, Phys. Rev. C 105, 035808 (2022), arXiv:2110.14896 [nucl-th]

2022 arXiv
[34]

Huth et al., Nature 606, 276 (2022), arXiv:2107.06229 [nucl-th]

S. Huth et al., Nature 606, 276 (2022), arXiv:2107.06229 [nucl-th]

2022 arXiv
[35]

I. Tews, R. Somasundaram, D. Lonardoni, H. G¨ ottling, R. Seutin, J. Carlson, S. Gandolfi, K. Hebeler, and A. Schwenk, Phys. Rev. Res. 7, 033024 (2025), arXiv:2407.08979

2025 arXiv
[36]

F. Alp, Y. Dietz, K. Hebeler, and A. Schwenk, Phys. Rev. C 112, 055802 (2025), arXiv:2504.18259. 26 5 10 15 20 25 30 35E/A [MeV] (a) Meta-model Smooth Spline 0.02 0.03 0.04 0.05 0.06 xβ [—] (b) 0.05 0.10 0.15 0.20 0.25 0.30 Density n [fm−3] 0 2 4 6 8 10 12p [MeVfm−3] (c) Inter...

2025
[37]

Drischler, V

C. Drischler, V. Som` a, and A. Schwenk, Phys. Rev. C 89, 025806 (2014), arXiv:1310.5627

2014 arXiv
[38]

Drischler, K

C. Drischler, K. Hebeler, and A. Schwenk, Phys. Rev. C 93, 054314 (2016), arXiv:1510.06728

2016 arXiv
[39]

Wellenhofer, J

C. Wellenhofer, J. W. Holt, and N. Kaiser, Phys. Rev. C 93, 055802 (2016), arXiv:1603.02935

2016 arXiv
[40]

Somasundaram, C

R. Somasundaram, C. Drischler, I. Tews, and J. Margueron, Phys. Rev. C 103, 045803 (2021), arXiv:2009.04737

2021 arXiv
[41]

Wen and J

P. Wen and J. W. Holt, Phys. Rev. C103, 064002 (2021), arXiv:2012.02163 [nucl-th]

2021 arXiv
[42]

Keller, C

J. Keller, C. Wellenhofer, K. Hebeler, and A. Schwenk, Phys. Rev. C 103, 055806 (2021), arXiv:2011.05855

2021 arXiv
[43]

Keller, K

J. Keller, K. Hebeler, and A. Schwenk, Phys. Rev. Lett. 130, 072701 (2023), arXiv:2204.14016 [nucl-th]

2023 arXiv
[44]

D. J. C. MacKay, Information Theory, Inference, and Learning Algorithms (Cambridge University Press, Cam- bridge, UK, 2003)

2003
[45]

C. E. Rasmussen and C. K. I. Williams, Gaussian Pro- cesses for Machine Learning , Adaptive computation and machine learning series (University Press Group Limited, Cambridge, MA, 2006)

2006
[46]

K. P. Murphy, Machine Learning: A Probabilistic Per- spective (The MIT Press, 2012)

2012
[47]

Gelman, J

A. Gelman, J. Carlin, H. Stern, D. Dunson, A. Vehtari, and D. Rubin, Bayesian Data Analysis, Third Edition , Chapman & Hall/CRC Texts in Statistical Science (Tay- lor & Francis, 2013)

2013
[48]

R. M. Neal, Priors for infinite networks, in Bayesian Learning for Neural Networks (Springer New York, New York, NY, 1996) pp. 29–53

1996
[49]

C. K. I. Williams, in NIPS (1996)

1996
[50]

J. Lee, Y. Bahri, R. Novak, S. S. Schoenholz, J. Pen- nington, and J. Sohl-Dickstein, Deep neural networks as gaussian processes (2018), arXiv:1711.00165 [stat.ML]

2018 arXiv
[51]

A. G. de G. Matthews, M. Rowland, J. Hron, R. E. Turner, and Z. Ghahramani, Gaussian process behaviour in wide deep neural networks (2018), arXiv:1804.11271 [stat.ML]

2018 arXiv
[52]

S. A. Sundberg and R. J. Furnstahl, J. Phys. G 52, 115106 (2025), arXiv:2508.01001 [nucl-th]

2025 arXiv
[53]

Drischler, R

C. Drischler, R. J. Furnstahl, J. A. Melendez, and D. R. Phillips, Phys. Rev. Lett. 125, 202702 (2020), arXiv:2004.07232 [nucl-th]. 27 0.04 0.05 0.07 0.08 0.10 0.11 0.12 0.14 0.15 n [fm−3] 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040 0.045 0.050 0.055 0.060x [—] Interaction...

2020 arXiv
[54]

Drischler, J

C. Drischler, J. A. Melendez, R. J. Furnstahl, and D. R. Phillips, Phys. Rev. C 102, 054315 (2020), arXiv:2004.07805 [nucl-th]

2020 arXiv
[55]

Svensson, A

I. Svensson, A. Tichai, K. Hebeler, and A. Schwenk, Phys. Rev. C 113, 024303 (2026), arXiv:2507.09079 [nucl-th]

2026
[56]

G¨ ottling, L

H. G¨ ottling, L. Hoff, K. Hebeler, and A. Schwenk, Phys. Rev. C 113, 055804 (2026), arXiv:2512.19593

2026 arXiv
[57]

A. C. Semposki, C. Drischler, R. J. Furnstahl, J. A. Melendez, and D. R. Phillips, Phys. Rev. C 111, 035804 (2025), arXiv:2404.06323 [nucl-th]

2025 arXiv
[58]

A. C. Semposki, C. Drischler, R. J. Furnstahl, and D. R. Phillips, Phys. Rev. C 113, 015808 (2026), arXiv:2505.18921 [nucl-th]

2026 arXiv
[59]

Gorda, O

T. Gorda, O. Komoltsev, A. Kurkela, and J. Schaffner- Bielich, (2026), arXiv:2606.23929

2026 arXiv
[60]

Legred and N

I. Legred and N. Yunes, (2026), arXiv:2603.26973

2026
[61]

Finch, I

E. Finch, I. Legred, K. Chatziioannou, R. Essick, S. Han, and P. Landry, Phys. Rev. D 112, 103023 (2025), arXiv:2505.13691

2025
[62]

Gorda, O

T. Gorda, O. Komoltsev, A. Kurkela, and E. Sunde, Astrophys. J. 1002, 40 (2026), arXiv:2512.18044 [astro- ph.HE]

2026 arXiv
[63]

Legred, L

I. Legred, L. Brodie, A. Haber, R. Essick, and K. Chatzi- ioannou, (2025), arXiv:2505.07677

2025
[64]

S. Ng, I. Legred, L. Suleiman, P. Landry, L. Traylor, and J. Read, Class. Quant. Grav. 42, 205008 (2025), arXiv:2507.03232

2025
[65]

Golomb, I

J. Golomb, I. Legred, K. Chatziioannou, A. Abac, and T. Dietrich, Phys. Rev. D 110, 063014 (2024), arXiv:2403.07697

2024
[66]

Essick, I

R. Essick, I. Legred, K. Chatziioannou, S. Han, and P. Landry, Phys. Rev. D 108, 043013 (2023), arXiv:2305.07411

2023 arXiv
[67]

Legred, B

I. Legred, B. O. Sy-Garcia, K. Chatziioannou, and R. Es- sick, Phys. Rev. D 109, 023020 (2024), arXiv:2310.10854

2024 arXiv
[68]

Legred, K

I. Legred, K. Chatziioannou, R. Essick, S. Han, and P. Landry, Phys. Rev. D 104, 063003 (2021), arXiv:2106.05313

2021 arXiv
[69]

Landry and R

P. Landry and R. Essick, Phys. Rev. D99, 084049 (2019), arXiv:1811.12529

2019 arXiv
[70]

Duvenaud, Automatic model construction with Gaus- sian processes, Ph.D

D. Duvenaud, Automatic model construction with Gaus- sian processes, Ph.D. thesis, Apollo - University of Cam- bridge Repository
[71]

Mroczek, M

D. Mroczek, M. C. Miller, J. Noronha-Hostler, and N. Yunes, Phys. Rev. D 110, 123009 (2024), arXiv:2309.02345 [astro-ph.HE]

2024 arXiv
[72]

Bradbury, R

J. Bradbury, R. Frostig, P. Hawkins, M. J. Johnson, C. Leary, S. Kind, L. Martens, A. Passos, S. Gross, S. Saxena, V. Saladi, J. Lim, M. van den Bosch, J. Burnim, and A. Paszke, JAX: Composable transfor- mations of Python+NumPy programs (2018)

2018
[73]

Chilenski, M

M. Chilenski, M. Greenwald, Y. Marzouk, N. Howard, A. White, J. Rice, and J. Walk, Nucl. Fusion 55, 023012 (2015)

2015
[74]

Piekarewicz and M

J. Piekarewicz and M. Centelles, Phys. Rev. C79, 054311 (2009), arXiv:0812.4499

2009 arXiv
[75]

Piekarewicz, J

J. Piekarewicz, J. Phys. G 37, 064038 (2010), arXiv:0912.5103

2010 arXiv
[76]

J. A. Melendez, R. J. Furnstahl, D. R. Phillips, M. T. Pratola, and S. Wesolowski, Phys. Rev. C 100, 044001 (2019), arXiv:1904.10581

2019 arXiv
[77]

M. G. Genton, J. Mach. Learn. Res. 2, 299–312 (2002)

2002
[78]

Herlands, A

W. Herlands, A. Wilson, H. Nickisch, S. Flaxman, D. Neill, W. van Panhuis, and E. Xing, arXiv e-prints , arXiv:1511.04408 (2015), arXiv:1511.04408

2015 arXiv
[79]

A. G. Wilson, D. A. Knowles, and Z. Ghahramani, arXiv:1110.4411 (2011)

2011 arXiv
[80]

Pedregosa, G

F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cour- napeau, M. Brucher, M. Perrot, and E. Duchesnay, J. Mach. Learn. Res. 12, 2825–2830 (2011)

2011
[81]

Babuschkin, K

DeepMind, I. Babuschkin, K. Baumli, A. Bell, S. Bhu- patiraju, J. Bruce, P. Buchlovsky, D. Budden, T. Cai, A. Clark, I. Danihelka, A. Dedieu, C. Fantacci, J. God- win, C. Jones, R. Hemsley, T. Hennigan, M. Hes- sel, S. Hou, S. Kapturowski, T. Keck, I. Kemaev, M. King, M. Kunes...

2020
[82]

Y. G. Lee, Quantifying Correlated Uncertainties in Mi- croscopic Nuclear Equation of State Calculations with Gaussian Processes, Ph.D. dissertation, Ohio University, Athens, OH, USA (2026), in prep

2026
[83]

Hebeler, Phys

K. Hebeler, Phys. Rept. 890, 1 (2021), arXiv:2002.09548 [nucl-th]

2021 arXiv
[84]

Simonis, S

J. Simonis, S. R. Stroberg, K. Hebeler, J. D. Holt, and A. Schwenk, Phys. Rev. C 96, 014303 (2017), arXiv:1704.02915

2017 arXiv
[85]

Stroberg, J

S. Stroberg, J. Holt, A. Schwenk, and J. Simonis, Phys. Rev. Lett. 126, 022501 (2021), arXiv:1905.10475 [nucl- th]

2021 arXiv
[86]

Miyagi, S

T. Miyagi, S. R. Stroberg, P. Navr´ atil, K. Hebeler, and J. D. Holt, Phys. Rev. C 105, 014302 (2022), arXiv:2104.04688. 28 ncc = 0.080+0.036 −0.021 0.0250.0300.0350.040 xcc xcc = 0.0313+0.0093 −0.0063 5.66.06.46.87.2 E/Acc E/Acc = 6.65+0.66 −0.57 0.30 0.45 0.60 0.75 pcc pcc =...

2022 arXiv
[87]

Arthuis, K

P. Arthuis, K. Hebeler, and A. Schwenk, (2024), arXiv:2401.06675

2024 arXiv
[88]

Bonaiti, G

F. Bonaiti, G. Hagen, and T. Papenbrock, (2025), arXiv:2508.14217 [nucl-th]

2025 arXiv
[89]

Hebeler, S

K. Hebeler, S. K. Bogner, R. J. Furnstahl, A. Nogga, and A. Schwenk, Phys. Rev. C 83, 031301(R) (2011), arXiv:1012.3381

2011 arXiv
[90]

J. M. Maldonado, C. Drischler, R. J. Furnstahl, and P. Mlinari´ c, Phys. Rev. C 112, 024002 (2025), arXiv:2504.06092

2025 arXiv
[91]

H. D. Patterson and R. Thompson, Biometrika 58, 545 (1971)

1971
[92]

D. A. Harville, Journal of the American Statistical As- sociation 72, 320 (1977)

1977
[93]

D. A. Harville, Biometrika 61, 383 (1974)

1974
[94]

Roca-Maza and N

X. Roca-Maza and N. Paar, Prog. Part. Nucl. Phys. 101, 96 (2018), arXiv:1804.06256

2018 arXiv
[95]

T. Li, U. Garg, Y. Liu, R. Marks, B. K. Nayak, P. V. M. Rao, M. Fujiwara, H. Hashimoto, K. Kawase, K. Nakan- ishi, S. Okumura, M. Yosoi, M. Itoh, M. Ichikawa, R. Matsuo, T. Terazono, M. Uchida, T. Kawabata, H. Akimune, Y. Iwao, T. Murakami, H. Sakaguchi, S. Terashima, Y. Yasud...

2007 arXiv
[96]

Li et al

T. Li et al. , Phys. Rev. C 81, 034309 (2010), arXiv:1002.0896

2010 arXiv
[97]

Patel, U

D. Patel, U. Garg, M. Fujiwara, H. Akimune, G. Berg, M. Harakeh, M. Itoh, T. Kawabata, K. Kawase, B. Nayak, T. Ohta, H. Ouchi, J. Piekarewicz, M. Uchida, H. Yoshida, and M. Yosoi, Phys. Lett. B718, 447 (2012), arXiv:1209.0681

2012 arXiv
[98]

J. R. Stone, N. J. Stone, and S. A. Moszkowski, Phys. Rev. C 89, 044316 (2014), arXiv:1404.0744

2014 arXiv
[99]

D. N. Basu, P. Roy Chowdhury, and C. Samanta, Phys. Rev. C 80, 057304 (2009), arXiv:0907.5350

2009 arXiv
[100]

Bonaiti, A

F. Bonaiti, A. Porro, S. Bacca, A. Schwenk, and A. Tichai, Phys. Rev. C 113, 024333 (2026), arXiv:2511.15836

2026
[101]

Kaiser, Phys

N. Kaiser, Phys. Rev. C 91, 065201 (2015), arXiv:1504.00604

2015 arXiv
[102]

Virtanen, R

P. Virtanen, R. Gommers, T. E. Oliphant, M. Haber- land, T. Reddy, D. Cournapeau, E. Burovski, P. Pe- terson, W. Weckesser, J. Bright, S. J. van der Walt, M. Brett, J. Wilson, K. J. Millman, N. Mayorov, A. R. J. Nelson, E. Jones, R. Kern, E. Larson, C. J. Carey, ˙I. Po- lat, Y...

2020
[103]

Ravenhall, C

D. Ravenhall, C. Pethick, and J. Wilson, Physical Review Letters 50, 2066 (1983)

1983
[104]

Chamel and P

N. Chamel and P. Haensel, Living Reviews in relativity 11, 1 (2008)

2008
[105]

Grams, J

G. Grams, J. Margueron, R. Somasundaram, and S. Reddy, Few Body Syst. 62, 116 (2021), arXiv:2109.11857

2021 arXiv
[106]

Carreau, F

T. Carreau, F. Gulminelli, and J. Margueron, Eur. Phys. J. A 55, 188 (2019), arXiv:1902.07032

2019 arXiv
[107]

Lim and J

Y. Lim and J. W. Holt, Phys. Rev. C 95, 065805 (2017), arXiv:1702.02898

2017 arXiv
[108]

Tews, Phys

I. Tews, Phys. Rev. C 95, 015803 (2017), arXiv:1607.06998

2017 arXiv
[109]

B. Link, R. I. Epstein, and J. M. Lattimer, Physical Review Letters 83, 3362 (1999)

1999
[110]

J. M. Lattimer and M. Prakash, Physics reports 442, 109 (2007)

2007
[111]

J. W. Gibbs, American Journal of Science s3-16, 441 (1878)
[112]

Kubis, Physical Review C—Nuclear Physics 76, 025801 (2007)

S. Kubis, Physical Review C—Nuclear Physics 76, 025801 (2007)

2007
[113]

Kunjipurayil and J

A. Kunjipurayil and J. Piekarewicz, (2026), arXiv:2606.10090

2026 arXiv
[114]

Ducoin, P

C. Ducoin, P. Chomaz, and F. Gulminelli, Nuclear Physics A 789, 403 (2007)

2007
[115]

Xu, L.-W

J. Xu, L.-W. Chen, B.-A. Li, and H.-R. Ma, Phys. Rev. C 79, 035802 (2009), arXiv:0807.4477

2009 arXiv
[116]

Ducoin, J

C. Ducoin, J. Margueron, C. Providˆ encia, and I. Vidana, Physical Review C—Nuclear Physics 83, 045810 (2011)

2011
[117]

Xu, L.-W

J. Xu, L.-W. Chen, B.-A. Li, and H.-R. Ma, Astrophys. J. 697, 1549 (2009), arXiv:0901.2309

2009 arXiv
[118]

Ducoin, J

C. Ducoin, J. Margueron, and C. Providencia, EPL (Europhysics Letters) 91, 32001 (2010)

2010
[119]

Carreau, F

T. Carreau, F. Gulminelli, and J. Margueron, The Euro- pean Physical Journal A 55, 188 (2019)

2019
[120]

Li and M

B.-A. Li and M. Magno, Physical Review C 102, 045807 (2020)

2020
[121]

B. A. Brown et al. , J. Phys. G 52, 050501 (2025), arXiv:2410.06144

2025 arXiv
[122]

Somasundaram, C

R. Somasundaram, C. L. Armstrong, P. Giuliani, K. God- bey, S. Gandolfi, and I. Tews, Phys. Lett. B 866, 139558 (2025), arXiv:2404.11566 [nucl-th]

2025 arXiv
[123]

C. L. Armstrong, P. Giuliani, K. Godbey, R. Somasun- daram, and I. Tews, Phys. Rev. Lett.135, 142501 (2025), arXiv:2502.03680 [nucl-th]

2025
[124]

Curry, K

R. Curry, K. Hebeler, S. Gandolfi, A. Gezerlis, A. Schwenk, R. Somasundaram, and I. Tews, (2025), arXiv:2510.15860

2025
[125]

P. Cook, K. Yu, C. Drischler, and S. K. Bogner, (2026), arXiv:2607.19773

2026 arXiv
[126]

H¨ uther, K

T. H¨ uther, K. Vobig, K. Hebeler, R. Machleidt, and R. Roth, Phys. Lett. B 808, 135651 (2020), arXiv:1911.04955

2020 arXiv
[127]

Reinert, H

P. Reinert, H. Krebs, and E. Epelbaum, Eur. Phys. J. A 54, 86 (2018), arXiv:1711.08821

2018 arXiv
[128]

Epelbaum, H

E. Epelbaum, H. Krebs, and P. Reinert, Semi-local Nu- clear Forces From Chiral EFT: State-of-the-Art and Challenges, in Handbook of Nuclear Physics , edited by I. Tanihata, H. Toki, and T. Kajino (Springer Singapore,
[129]

Nosyk, D

Y. Nosyk, D. R. Entem, and R. Machleidt, Phys. Rev. C 104, 054001 (2021), arXiv:2107.06452 [nucl-th]

2021 arXiv
[130]

Piarulli and I

M. Piarulli and I. Tews, Front. in Phys. 7, 245 (2020), arXiv:2002.00032 [nucl-th]

2020 arXiv
[131]

W. G. Jiang, A. Ekstr¨ om, C. Forss´ en, G. Hagen, G. R. Jansen, and T. Papenbrock, Phys. Rev. C 102, 054301 (2020), arXiv:2006.16774 [nucl-th]

2020 arXiv
[132]

github.io/software/

BUQEYE collaboration, (2022), https://buqeye. github.io/software/

2022
[133]

Drischler, A

C. Drischler, A. Carbone, K. Hebeler, and A. Schwenk, Phys. Rev. C 94, 054307 (2016), arXiv:1608.05615

2016 arXiv
[134]

Marino, W

F. Marino, W. Jiang, and S. J. Novario, Phys. Rev. C 110, 054322 (2024), arXiv:2407.17098 [nucl-th]

2024
[135]

Foreman-Mackey, The Journal of Open Source Soft- ware 1, 24 (2016)

D. Foreman-Mackey, The Journal of Open Source Soft- ware 1, 24 (2016)

2016
[136]

Los Alamos National Laboratory (LANL), dfdjaxGP: JAX-based Gaussian Process tools for DFD applications, https://github.com/lanl/dfdjaxGP (2023), GitHub repository, accessed: 2026-03-20. 30

2023
[137]

(2024) https://n3as.berkeley.edu/

N3AS Collab. (2024) https://n3as.berkeley.edu/

2024
[138]

van Breugel, Y

F. van Breugel, Y. Liu, B. W. Brunton, and J. N. Kutz, PyNumDiff: A python package for numerical differentia- tion of noisy time-series data (2022), version 0.1.2.4

2022
[139]

Kluyver, B

T. Kluyver, B. Ragan-Kelley, F. P´ erez, B. Granger, M. Bussonnier, J. Frederic, K. Kelley, J. Hamrick, J. Grout, S. Corlay, P. Ivanov, D. Avila, S. Abdalla, C. Willing, and J. development team, in Positioning and Power in Academic Publishing: Players, Agents and Agendas, edit...

2016
[140]

J. D. Hunter, Computing in Science & Engineering 9, 90 (2007)

2007
[141]

A. G. d. G. Matthews, M. van der Wilk, T. Nickson, K. Fujii, A. Boukouvalas, P. Le´ on-Villagr´ a, Z. Ghahra- mani, and J. Hensman, Journal of Machine Learning Research 18, 1 (2017)

2017
[142]

C. R. Harris, K. J. Millman, S. J. van der Walt, R. Gom- mers, P. Virtanen, D. Cournapeau, E. Wieser, J. Taylor, S. Berg, N. J. Smith, R. Kern, M. Picus, S. Hoyer, M. H. van Kerkwijk, M. Brett, A. Haldane, J. F. del R ´ ıo, M. Wiebe, P. Peterson, P. G´ erard-Marchant, K. Shep-...

2020
[143]

Ambikasaran, D

S. Ambikasaran, D. Foreman-Mackey, L. Greengard, D. W. Hogg, and M. O’Neil, IEEE Transactions on Pat- tern Analysis and Machine Intelligence 38, 252 (2015), arXiv:1403.6015

2015 arXiv
[144]

The GPy authors, GPy: A Gaussian process frame- work in Python, http://github.com/SheffieldML/GPy (2012)

2012
[145]

van der Wilk, V

M. van der Wilk, V. Dutordoir, S. John, A. Artemev, V. Adam, and J. Hensman, arXiv:2003.01115 (2020)

2020 arXiv
[146]

Pinder and D

T. Pinder and D. Dodd, Journal of Open Source Software 7, 4455 (2022)

2022
[147]

Muyskens, B

A. Muyskens, B. Priest, I. Goumiri, and M. Schneider, arXiv:2104.14581 (2021)

2021 arXiv
[148]

J. R. Gardner, G. Pleiss, D. Bindel, K. Q. Weinberger, and A. G. Wilson, arXiv:1809.11165 (2018)

2018 arXiv
[149]

Binois and R

M. Binois and R. B. Gramacy, Journal of Statistical Software 98, 1–44 (2021)

2021
[2022]

1–25, arXiv:2206.07072 [nucl-th]

pp. 1–25, arXiv:2206.07072 [nucl-th]

arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

Ekstr¨ om, G

A. Ekstr¨ om, G. R. Jansen, K. A. Wendt, G. Hagen, T. Papenbrock, B. D. Carlsson, C. Forss´ en, M. Hjorth- Jensen, P. Navr´ atil, and W. Nazarewicz, Phys. Rev. C 91, 051301(R) (2015), arXiv:1502.04682

arXiv 2015

[2] [2]

Ekstr¨ om, G

A. Ekstr¨ om, G. Hagen, T. D. Morris, T. Papenbrock, and P. D. Schwartz, Phys. Rev. C 97, 024332 (2018), arXiv:1707.09028

arXiv 2018

[3] [3]

Drischler, K

C. Drischler, K. Hebeler, and A. Schwenk, Phys. Rev. Lett. 122, 042501 (2019), arXiv:1710.08220

arXiv 2019

[4] [4]

We can therefore determine the common mean process by conditioning on the observed sample mean

Posterior of the common mean EOS With the deviation kernel now calibrated, all compo- nents of the ensemble mean covariance are specified. We can therefore determine the common mean process by conditioning on the observed sample mean. At arbitrary evaluation points, η | ¯yt, xt, x∗ ∼ N mη,∗|t, Kη,∗∗|t , (46) with posterior mean and posterior covariance ma...

[5] [5]

Meta-model

Predictive distribution for a new EOS We now return to the predictive target in Eq. (26). Before conditioning on the MBPT data for the EOS, f (new)(x) is marginally distributed as f (new)(x) ∼ GP(µη(x), kη(x, x′) + kδ(x, x′)) , (51) corresponding to Eq. (31) when also the numerical noise is marginalized over and with the kernels defined in Eqs.(40) and (4...

arXiv

[6] [6]

W. G. Jiang, C. Forss´ en, T. Dj¨ arv, and G. Hagen, Phys. Rev. C 109, 064314 (2024), arXiv:2212.13216

arXiv 2024

[7] [7]

W. G. Jiang, C. Forss´ en, T. Dj¨ arv, and G. Hagen, Phys. Rev. C 109, L061302 (2024), arXiv:2212.13203 [nucl-th]. 24 n0 = 0.170+0.062 −0.038 −21 −18 −15 −12 −9 E0/A E0/A =−15.57+3.04 −4.70 150 200 250 300 350 K K = 223+116 −68 28 32 36 40 Sv Sv = 32.8+7.9 −5.1 40 50 60 70 L L = 51.0+18.7 −11.4 0.120.150.180.210.24 n0 −750−600−450−300−150 Kτ −21−18−15−12−...

arXiv 2024

[8] [8]

Drischler, P

C. Drischler, P. G. Giuliani, S. Bezoui, J. Piekarewicz, and F. Viens, Phys. Rev. C 110, 044320 (2024), arXiv:2405.02748

arXiv 2024