Pith. sign in

REVIEW 4 major objections 6 minor 94 references

Probing the refined performance of the Categorical-Boosting algorithm to the Hartree-Fock-Bogoliubov mass model with different Skyrme forces

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper reports that a CatBoost machine-learning refinement of six Skyrme-HFB nuclear mass tables cuts test-set errors to roughly 0.2 MeV and generalizes to nuclei measured after the training data.

desk verdict A competent, incremental CatBoost application to six Skyrme-HFB mass tables with solid ~0.2 MeV interpolation accuracy, but the 'generalization' claim needs tightening and the independence claim is wrong. read the letter →

arxiv 2505.10750 v1 pith:TFPGO74J submitted 2025-05-15 nucl-th

Jin-Liang Guo , Hua-Lei Wang , Zhen-Zhen Zhang , Min-Liang Liu This is my paper · ORCID
classification nucl-th
keywords nuclearmassmachinelearningCatBoostSkyrmeforceHartree-Fock-Bogoliubovbindingenergymodelrefinementgeneralization
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

The paper sets out to show that a gradient-boosting algorithm called CatBoost can repair the binding-energy predictions of six Skyrme-Hartree-Fock-Bogoliubov mass tables, bringing test-set root-mean-square errors from 1.4-7.0 MeV down to roughly 0.2 MeV. The authors train CatBoost on the difference between each theoretical mass and the experimental mass in the 2020 Atomic Mass Evaluation, using seven features based on proton number, neutron number, parity, and distance to magic numbers. They report that every refined model reaches about 0.2 MeV test error, reduces large model bias, and still agrees with 21 masses measured after the training data. The practical interest is that microscopic mean-field mass tables, often too inaccurate for nuclear-structure or astrophysical applications, could be upgraded as a package rather than re-fitted.

What carries the argument

The load-bearing object is the mass residual $\delta(Z,N)=B_{\rm th}(Z,N)-B_{\rm exp}(Z,N)$ between a Skyrme-HFB table and experiment, together with the CatBoost gradient-boosting algorithm that learns $\delta$ from seven features: $Z$, $N$, $N/Z$, even-odd flags ${\rm Zeo},{\rm Neo}$, and distances $|Z-m|$, $|N-m|$ to magic numbers. CatBoost's ordered boosting replaces standard gradient estimates to reduce prediction shift; the refined mass is then $B_{\rm ml}=B_{\rm th}-\delta_{\rm learned}$. The claimed repair is quantified by a model-repair coefficient $R_{\rm MR}=(\sigma_{\rm HFB}-\sigma_{\rm test})/\sigma_{\rm HFB}$, which exceeds 84% for all six forces.

What would settle it

Train the CatBoost-refined model on nuclei in one region, say $Z<50$, and test on heavier nuclei with a similar split, or evaluate the 21 newly measured nuclei after excluding all nuclei in their isotopic chains from training; if the rms error jumps far above 0.2 MeV or fails to beat the bare HFB table, the smooth-residual assumption is falsified.

Watch

Extended reading notes

Core claim

For each of six Skyrme parameter sets (SkM*, SkP, SLy4, SV-min, UNEDF0, UNEDF1), the paper's central claim is that the residual $B_{\rm HFB}(Z,N)-B_{\rm exp}(Z,N)$ is a learnable function of the seven-feature set and that CatBoost can learn it from the measured nuclei with $Z,N\ge 8$. After hyperparameter tuning with ten-fold cross-validation, the refined models reach testing-set rms deviations of 0.189-0.259 MeV, model-repair coefficients above 84%, and residuals that become visibly more random across the nuclear chart. On 21 nuclei measured after AME2020, the refined models give rms deviations of 0.137-0.293 MeV. The authors also claim that the best bare Skyrme force (UNEDF0, 1.43 MeV) is not the best refined force, indicating that the algorithm captures different missing physics for different interactions.

Load-bearing premise

The load-bearing premise is that the difference between a Skyrme-HFB binding energy and the measured value varies smoothly with proton number, neutron number, parity, and distance to magic numbers, so patterns learned on measured nuclei carry over to unmeasured and extrapolated nuclei; if that difference has random or local structure the features cannot capture, the claimed accuracy would not persist.

Editorial extensions

If this is right

  • All six adopted Skyrme-HFB tables can be repaired to roughly 0.2 MeV test accuracy, bringing microscopic mass models closer to the level needed for astrophysical applications.
  • The refined models generalize to 21 nuclei measured after AME2020, with rms deviations of 0.137-0.293 MeV, so the correction is not limited to nuclei already in the training set.
  • The residual distributions become more random and the large model bias decreases, indicating that the algorithm is capturing part of the missing physics rather than only memorizing the training data.
  • The ranking of Skyrme forces by predictive power changes after refinement: UNEDF0 is best in the bare HFB calculations, but UNEDF1 gives the best refined test error, so model selection should be based on refined rather than bare performance.
  • Extrapolation to drip-line nuclei worsens with distance from the training region, but the refined predictions often remain better than the bare HFB predictions except for the most distant proton-rich cases.

Reading between the lines

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

  • A testable extension is to treat CatBoost as a universal repair layer: if the residual remains smooth for other mean-field mass tables, the same seven-feature pipeline should produce comparable roughly 0.2 MeV test errors without changing the physics model.
  • The reversal of the best-force ranking after refinement suggests model selection should be done on refined, not bare, predictions; retraining the six models when the next atomic mass evaluation appears could check which Skyrme force then yields the best extrapolation.
  • Because the refined masses are smooth in $Z$ and $N$, they can be used to refit the coefficients of simple liquid-drop-type mass formulas; the paper already does this, and the same trick could generate improved constraints on Skyrme energy-density functionals.
  • The near-0.2 MeV test floor, larger than the roughly 25 keV experimental uncertainty, may reflect irreducible model randomness; one can probe this by testing whether the error saturates as training data grow, which would indicate how much missing physics is recoverable from present features.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper applies the CatBoost gradient-boosting algorithm to refine six Skyrme-HFB mass tables (SkM*, SkP, SLy4, SV-min, UNEDF0, UNEDF1) by learning the residual delta(Z,N) = B_HFB - B_exp from AME2020 data. Seven nuclear features are used (Table I), eight feature combinations are compared (Table II), and the M8 combination is selected using 10-fold cross-validation. After a grid search over hyperparameters (Table III), the reported test-set rms values are 0.189-0.259 MeV, with model-repair coefficients RMR = (sigma_HFB - sigma_test)/sigma_HFB between 84.5% and 96.3% (Table V). The authors also test 21 nuclei outside AME2020, reporting post-refinement rms values around 0.14-0.29 MeV, and examine extrapolation to drip-line nuclei in Fig. 16. They conclude that CatBoost can 'pick up the missing physics' and that the refined predictions are nearly independent of the bare model, while conceding in the Summary that extrapolation in very extreme regions is unreliable.

Significance. If the 0.2 MeV refined accuracy and the generalization claim held, this would be a practically useful contribution to nuclear mass prediction, with potential impact on r-process abundance calculations and the interpretation of new mass measurements. The paper's strengths are the systematic six-force comparison, the repeated random-split evaluation with 500 runs and uncertainty estimates, the use of external AME2020 data, and the out-of-sample check with 21 newly measured nuclei. The machine-checked style of the numerical tables is a useful feature. However, the central generalization claim is currently overstated: the evaluation protocol does not provide a fully independent test set for feature and hyperparameter selection, and the extrapolation evidence is either small in sample size (21 nuclei) or shows performance degradation with distance (Fig. 16). The interpolation performance is solid, but the extrapolation claim needs additional support before the paper can be accepted.

major comments (4)
  1. [Section III, Figs. 3, 5, 8 and Table IV] The evaluation protocol does not separate model selection from final testing. The M8 feature set is chosen using 10-fold cross-validation on the full dataset (Fig. 3), and the optimal hyperparameters in Table IV are selected by a further 10-fold CV on the same full dataset. The reported test-set rms values in Table V are then obtained from random 4:1 splits of that same dataset. This means the hyperparameters and feature set may have been partially tuned on nuclei that later appear in the test folds, so the 0.189-0.259 MeV test values are not fully independent of model selection. Please use a nested cross-validation scheme or hold out a completely untouched test set for the final evaluation, and report how the test rms changes when feature/hyperparameter selection is performed only on training folds.
  2. [Section III, Fig. 16 and Section IV] The abstract and the Summary claim 'good generalization abilities', but the paper's own extrapolation test in Fig. 16 shows that the refined-model rms increases with extrapolation distance, and at the largest proton-rich distance (pi4) the CatBoost-refined model is not better than the bare HFB model. The Summary itself concedes that 'in the very extreme nuclear regions ... the extrapolation performance of the machine learning will be unreliable'. Additionally, the 21-nucleus test in Fig. 15 is not accompanied by an analysis of how far those nuclei are, in the chosen feature space, from the training region; most of them lie near the stability line and are plausibly close neighbors of training nuclei. The claim of 'good generalization' should therefore be restricted to interpolation or supported by a quantitative feature-distance analysis and a clear statement of the distance at which performance degrades.
  3. [Section III, Fig. 14] The claim that the refined predictions 'hardly depend' on the bare HFB model is contradicted by the construction B_ml = B_th - delta, where delta is trained on B_th - B_exp. The residual is a function of the bare model by construction, so the weak Pearson correlation between the bare and refined rms values across 500 splits does not establish independence of the bare model. The correct statement is that the residual fit removes most of the large model bias, not that the refined predictions are nearly independent of the underlying physical model. Please rephrase the text around Fig. 14 accordingly.
  4. [Table V] The model-repair coefficient RMR = (sigma_HFB - sigma_test)/sigma_HFB is presented as a measure of the algorithm's 'model-repair ability'. However, sigma_HFB is computed over all 2457 nuclei while sigma_test is computed on a random 20% subset, and the test set is also used for early stopping during training. The reported RMR values of 84.5-96.3% are therefore optimistic in a way that should be acknowledged. If nested cross-validation is adopted as suggested above, the RMR values should be recomputed on the held-out folds only.
minor comments (6)
  1. [Abstract] There are several typos: 'the the Hartree-Fock-Bogoliubov', 'paraterer', 'accurancies', and 'Intrestingly'. These should be corrected.
  2. [Fig. 11 caption] The scaling factor F is defined as 1/(delta_max - delta_min), but it is not stated over which dataset (training set, testing set, or all data) the maximum and minimum residuals are taken.
  3. [Table III] The iteration domain [1000, 5000] with increment 2000 yields only 1000, 3000, and 5000. If 2000 and 4000 were also scanned, this should be stated; otherwise the domain/increment combination should be corrected.
  4. [Fig. 16] The vertical axis label in Fig. 16(b) is missing. The text refers to 'rms derivations', but the axis should explicitly state the quantity and its units.
  5. [Eq. (1)] The loss function L(y, F(x)) is introduced abstractly, but the actual metric used throughout the paper is the rms residual. For self-containedness, please state explicitly that the squared-error loss is used in the CatBoost training.
  6. [References] Reference [45] (Mass Explorer) is an online resource; please include an access date and, if possible, a version or a permanent identifier.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the CatBoost residual-learning evaluation is anchored to external AME2020 data and to genuinely held-out nuclei, with no load-bearing self-citation.

full rationale

The paper's central procedure is standard residual supervised learning: the target is the residual delta(Z,N) = B_HFB(Z,N) - B_exp(Z,N), and the refined mass is B_ML = B_HFB - delta_learned, where delta_learned is a CatBoost regression trained on seven features (Z, N, N/Z, Zeo, Neo, |Z-m|, |N-m|) chosen from the cited literature. This construction is not definitionally circular: the learned residual is a function fitted on a random 80% portion of AME2020, and the reported performance is measured on the held-out 20% of AME2020, on 21 nuclei measured after AME2020, and on drip-line nuclei explicitly removed from the training set in the extrapolation test of Fig. 16. None of these test quantities is used to define the training target or the fitted model, so the ~0.2 MeV test accuracy is an externally anchored empirical result rather than an identity built into the method. The model-repair coefficient RMR = (sigma_HFB - sigma_test)/sigma_HFB is a descriptive ratio computed after evaluation, not an input that forces the prediction. The only same-group citation, Ref. [37], is used for the shape of the extrapolation protocol only; the protocol is fully described in the text and its numerical results are produced by the present training/evaluation split, so the citation is not load-bearing. The interpretive phrase 'picking up missing physics' is a post-hoc characterization of the residual reduction, not an assumption used in deriving the numbers. Concerns about random splits being mostly interpolation, the small size of the 21-nucleus external set, and the degraded extrapolation at large distance are legitimate correctness and generalization-risk remarks, but they are not circularity: they do not show that any prediction is equivalent by construction to its input.

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

The paper introduces no new physical entities. It relies on experimental mass data, published HFB mass tables, a set of chosen nuclear features, and tuned CatBoost hyperparameters. The main inherited assumptions are the accuracy of AME2020, the validity of the Mass Explorer tables, and the learnability of the residual with the selected features.

free parameters (5)
  • CatBoost tree depth d = SkM*: 5, SkP: 5, SLy4: 6, SV-min: 6, UNEDF0: 4, UNEDF1: 6
    Chosen by grid search on validation folds; directly affects test rms.
  • CatBoost learning rate lr = SkM*: 0.06, SkP: 0.06, SLy4: 0.04, SV-min: 0.04, UNEDF0: 0.16, UNEDF1: 0.06
    Tuned in the hyperparameter grid; controls contribution of each tree to the ensemble.
  • CatBoost L2 regularization coefficient lambda_reg = 3 for SkM*, SkP, SLy4, SV-min, UNEDF1; 6 for UNEDF0
    Selected by grid search to reduce overfitting; listed in Table IV.
  • CatBoost iteration count (early stopping) = SkM*: 3916, SkP: 3078, SLy4: 4000, SV-min: 3834, UNEDF0: 1374, UNEDF1: 2496
    Determined by early stopping with patience 30; listed in Table IV.
  • Proton magic number used in feature |Z-m| beyond Z=82 = 126
    Selected to match Ref. [34]; the paper also tests 114 and 120 and finds no significant difference (Fig. 4).
assumptions (5)
  • domain assumption AME2020 experimental masses are accurate enough to serve as training labels and evaluation ground truth.
    The paper uses only nuclei with measured binding energies from AME2020 (Z,N>=8) as targets; if these masses were biased, the residual learning would inherit the bias.
  • domain assumption The HFB mass tables from the Mass Explorer (Ref. [45]) correctly correspond to the cited Skyrme forces SkM*, SkP, SLy4, SV-min, UNEDF0, UNEDF1.
    The paper does not recompute HFB tables; all residuals and refined masses depend on these tables.
  • domain assumption The residual between HFB and experiment is predominantly model deficiency (missing physics) rather than random model uncertainty.
    Stated in Section I and used to define the model-repair coefficient RMR; the ML correction cannot distinguish bias from noise.
  • ad hoc to paper The residual delta(Z,N) is a learnable function of the seven chosen features (Z, N, N/Z, Zeo, Neo, |Z-m|, |N-m|).
    Feature set is selected from prior literature and validation; no guarantee it captures all residual structure, especially at the drip line.
  • domain assumption Standard shell magic numbers (proton: 8,20,28,50,82,126; neutron: 8,20,28,50,82,126,184) are used in the shell-distance features.
    The paper tests only alternative proton magic numbers beyond 82; other shell closures are taken as fixed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Probing the refined performance of the Categorical-Boosting algorithm to the Hartree-Fock-Bogoliubov mass model with different Skyrme forces." pith.science (2026). https://pith.science/paper/TFPGO74J

@misc{pith2026250510750,
  author       = {Pith},
  title        = {Pith review of: Probing the refined performance of the Categorical-Boosting algorithm to the Hartree-Fock-Bogoliubov mass model with different Skyrme forces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TFPGO74J}},
  note         = {Machine review of arXiv:2505.10750}
}
read the original abstract

Nuclear mass can offer profound insights into many physical branches, e.g., nuclear physics and astrophysics, while the predicted accuracy by nuclear mass models is usually far from satisfactory until now, especially within the fully microscopic self-consistent mean-field theory. In this project, we present the predictive power for the binding energy within the the Hartree-Fock-Bogoliubov (HFB) methods with six widely used Skyrme forces (SkM*, SkP, SLy4, SV-min, UNEDF0 and UNEDF1) and evaluate the refined performance of the machine learning based on a novel Categorical Boosting (CatBoost) algorithm to the Skyrme HFB mass models. The root-mean-square (rms) deviations between the bare HFB calculations with different Skyrme forces and the available experimental data range from the minimum, about 1.43 MeV, for the UNEDF0 parameter set to the maximum, about 7.03 MeV, for the SkM* paraterer set. For the CatBoost-refined HFB predictions, the predictive power can be significantly improved. All the prediction accurancies on the testing set can reach the level around 0.2 MeV and, meanwhile, the large model bias can be reduced. The model-repair coefficients for the adopted Skyrme parameter sets are uniformly more than 80\%. Moreover, for 21 newly measured nuclei outside AME2020, the predicted masses by the CatBoost-refined HFB models are also in good agreement with the experimental data, illustrating their good generalization abilities. Intrestingly, it is found that the optimal Skyrme parameter set that possesses the highest predictive power for the bare HFB mass calculations may be not the best candidate for the CatBoost-refined HFB model, indicating the different abilities of picking up the missing ``physics'' for different Skyrme forces by the CatBoost algorithm.

Figures

Figures reproduced from arXiv: 2505.10750 by the authors.

Figure 1
Figure 1. FIG. 1. (Color online) Residual [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. (Color online) Similar to Fig [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. (Color online) With different Skyrme parameters, [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (13 more)
Figure 5
Figure 5. Figure 5: FIG. 5. (Color online) Distribution of the rms deviations ( [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 4
Figure 4. Figure 4: FIG. 4. (Color online) Contour plots of shell gaps in nuclei [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 7
Figure 7. Figure 7: FIG. 7. (Color online) The rms deviations ( [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. (Color online) Scatter plots of the rms deviations in [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. (Color online) Importance ranking for the input fea [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10. (Color online) Feature importance obtained accord [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: (a) to (d), with the increasing neutron number. In order to further compare the predicted accuracies in different nuclear domains, e.g., from light to heavy mass regions (similar to those in Refs. [75–77]), we di￾vided the the experimental dataset collected in AME2020…
Figure 12
Figure 12. Figure 12: FIG. 12. (Color online) (a) The rms deviations [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]
Figure 13
Figure 13. Figure 13: FIG. 13. (Color online) Similr to Fig [PITH_FULL_IMAGE:figures/full_fig_p011_13.png]
Figure 14
Figure 14. Figure 14: FIG. 14. (Color online) Two-dimensional scatter plots of the [PITH_FULL_IMAGE:figures/full_fig_p012_14.png]
Figure 16
Figure 16. Figure 16: FIG. 16. (Color online) (a) The selected nuclei in the training [PITH_FULL_IMAGE:figures/full_fig_p012_16.png]
Figure 15
Figure 15. Figure 15: FIG. 15. (Color online) Residuals [PITH_FULL_IMAGE:figures/full_fig_p012_15.png]
Figure 17
Figure 17. Figure 17: FIG. 17. (Color online) Residual [PITH_FULL_IMAGE:figures/full_fig_p013_17.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

94 extracted references · 72 canonical work pages

  1. [1]

    Lunney, J

    D. Lunney, J. M. Pearson, and C. Thibault, Recent trends in the determination of nuclear masses, Rev. Mod. Phys. 75, 1021 (2003)

    2003

  2. [2]

    E. M. Burbidge, G. R. Burbidge, W. A. Fowler, and F. Hoyle, Synthesis of the Elements in Stars, Rev. Mod. Phys. 29, 547 (1957)

    1957

  3. [3]

    C. Ma, M. Bao, Z. M. Niu, Y. M. Zhao, and A. Arima, New extrapolation method for predicting nuclear masses, Phys. Rev. C 101, 045204 (2020)

    2020

  4. [4]

    M. Wang, W. Huang, F. Kondev, G. Audi, and S. Naimi, The AME 2020 atomic mass evaluation (II). Tables, graphs and references, Chin. Phys. C 45, 030003 (2021)

    2021

  5. [5]

    C. F. v. Weizs¨ acker, Zur Theorie der Kernmassen, Z. Angew. Phys. 96, 431 (1935)

    1935

  6. [6]

    M¨ oller, W

    P. M¨ oller, W. D. Myers, H. Sagawa, and S. Yoshida, New Finite-Range Droplet Mass Model and Equation-of-State Parameters, Phys. Rev. Lett. 108, 052501 (2012)

    2012

  7. [7]

    N. Wang, M. Liu, X. Wu, and J. Meng, Surface diffuse- ness correction in global mass formula, Phys. Lett. B734, 215 (2014)

    2014

  8. [8]

    L. Geng, H. Toki, and J. Meng, Masses, Deformations and Charge Radii-Nuclear Ground-State Properties in the Relativistic Mean Field Model, Prog. Theor. Phys 113, 785 (2005)

    2005

Show all 94 references
  1. [9]

    Moller, J

    P. Moller, J. Nix, W. Myers, and W. Swiatecki, Nuclear Ground-State Masses and Deformations, At. Data Nucl. Data Tables 59, 185 (1995)

    1995

  2. [10]

    Goriely, N

    S. Goriely, N. Chamel, and J. M. Pearson, Further explo- rations of Skyrme-Hartree-Fock-Bogoliubov mass formu- las. XVI. Inclusion of self-energy effects in pairing, Phys. Rev. C 93, 034337 (2016)

    2016

  3. [11]

    Goriely, N

    S. Goriely, N. Chamel, and J. M. Pearson, Skyrme- Hartree-Fock-Bogoliubov Nuclear Mass Formulas: Cross- ing the 0.6 MeV Accuracy Threshold with Microscop- ically Deduced Pairing, Phys. Rev. Lett. 102, 152503 (2009)

    2009

  4. [12]

    X. Xia, Y. Lim, P. Zhao, H. Liang, X. Qu, Y. Chen, H. Liu, L. Zhang, S. Zhang, Y. Kim, and J. Meng, The limits of the nuclear landscape explored by the relativistic continuum Hartree–Bogoliubov theory, At. Data Nucl. Data Tables 121-122, 1 (2018)

    2018

  5. [13]

    Carleo, I

    G. Carleo, I. Cirac, K. Cranmer, L. Daudet, M. Schuld, N. Tishby, L. Vogt-Maranto, and L. Zdeborov´ a, Machine learning and the physical sciences, Rev. Mod. Phys. 91, 045002 (2019)

    2019

  6. [14]

    Boehnlein, M

    A. Boehnlein, M. Diefenthaler, N. Sato, M. Schram, V. Ziegler, C. Fanelli, M. Hjorth-Jensen, T. Horn, M. P. Kuchera, D. Lee, W. Nazarewicz, P. Ostroumov, K. Orginos, A. Poon, X.-N. Wang, A. Scheinker, M. S. Smith, and L.-G. Pang, Colloquium: Machine learning in nuclear physics...

    2022

  7. [15]

    Silver, A

    D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, G. van den Driessche, J. Schrittwieser, I. Antonoglou, V. Panneershelvam, M. Lanctot, S. Dieleman, D. Grewe, J. Nham, N. Kalchbrenner, I. Sutskever, T. Lillicrap, M. Leach, K. Kavukcuoglu, T. Graepel, and D. Hassabis, Mas...

    2016

  8. [16]

    Chang, T

    S. Chang, T. Cohen, and B. Ostdiek, What is the machine learning?, Phys. Rev. D 97, 056009 (2018)

    2018

  9. [17]

    Gernoth, J

    K. Gernoth, J. Clark, J. Prater, and H. Bohr, Neural net- work models of nuclear systematics, Phys. Lett. B 300, 1 (1993)

    1993

  10. [18]

    Athanassopoulos, E

    S. Athanassopoulos, E. Mavrommatis, K. Gernoth, and J. Clark, Nuclear mass systematics using neural net- works, Nucl. Phys. A 743, 222 (2004)

    2004

  11. [19]

    J. W. CLARK and H. LI, Application of support vector machines to global prediction of nuclear properties, Int. 15 J. Mod. Phys. B 20, 5015 (2006)

    2006

  12. [20]

    Ming, H.-F

    X.-C. Ming, H.-F. Zhang, R.-R. Xu, X.-D. Sun, Y. Tian, and Z.-G. Ge, Nuclear mass based on the multi-task learning neural network method, Nucl. Sci. Tech. 33, 48 (2022)

    2022

  13. [21]

    Li, C.-N

    C.-Q. Li, C.-N. Tong, H.-J. Du, and L.-G. Pang, Deep learning approach to nuclear masses and α-decay half- lives, Phys. Rev. C 105, 064306 (2022)

    2022

  14. [22]

    X.-K. Le, N. Wang, and X. Jiang, Nuclear mass pre- dictions with multi-hidden-layer feedforward neural net- work, Nucl. Phys. A 1038, 122707 (2023)

    2023

  15. [23]

    Zeng, Y.-Y

    L.-X. Zeng, Y.-Y. Yin, X.-X. Dong, and L.-S. Geng, Nu- clear binding energies in artificial neural networks, Phys. Rev. C 109, 034318 (2024)

    2024

  16. [24]

    Niu and H

    Z. Niu and H. Liang, Nuclear mass predictions based on Bayesian neural network approach with pairing and shell effects, Phys. Lett. B 778, 48 (2018)

    2018

  17. [25]

    Z. M. Niu and H. Z. Liang, Nuclear mass predictions with machine learning reaching the accuracy required by r-process studies, Phys. Rev. C 106, L021303 (2022)

    2022

  18. [26]

    M. Li, T. M. Sprouse, B. S. Meyer, and M. R. Mumpower, Atomic masses with machine learning for the astrophys- ical r process, Phys. Lett. B 848, 138385 (2024)

    2024

  19. [27]

    H. F. Zhang, L. H. Wang, J. P. Yin, P. H. Chen, and H. F. Zhang, Performance of the Levenberg–Marquardt neural network approach in nuclear mass prediction, J. Phys. G: Nucl. Part. Phys. 44, 045110 (2017)

    2017

  20. [28]

    Y¨ uksel, D

    E. Y¨ uksel, D. Soydaner, and H. Bahtiyar, Nuclear mass predictions using machine learning models, Phys. Rev. C 109, 064322 (2024)

    2024

  21. [29]

    Carnini and A

    M. Carnini and A. Pastore, Trees and forests in nuclear physics, J. Phys. G: Nucl. Part. Phys. 47, 082001 (2020)

    2020

  22. [30]

    Y¨ uksel, D

    E. Y¨ uksel, D. Soydaner, and H. Bahtiyar, Nuclear bind- ing energy predictions using neural networks: Applica- tion of the multilayer perceptron, Int. J. Mod. Phys. E 30, 2150017 (2021)

    2021

  23. [31]

    W. He, Q. Li, Y. Ma, Z. Niu, J. Pei, and Y. Zhang, Ma- chine learning in nuclear physics at low and intermedi- ate energies, Sci. China- Phys. Mech. Astron 66, 282001 (2023)

    2023

  24. [32]

    Prokhorenkova, G

    L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Doro- gush, and A. Gulin, CatBoost: unbiased boosting with categorical features (Curran Associates, Inc., 2018)

    2018

  25. [33]

    Zhang, Z

    Y. Zhang, Z. Zhao, and J. Zheng, CatBoost: A new ap- proach for estimating daily reference crop evapotranspi- ration in arid and semi-arid regions of Northern China, Journal of Hydrology 588, 125087 (2020)

    2020

  26. [34]

    Gao, Y.-J

    Z.-P. Gao, Y.-J. Wang, H.-L. L¨ u, Q.-F. Li, C.-W. Shen, and L. Liu, Machine learning the nuclear mass, Nucl. Sci. Tech. 32, 109 (2021)

    2021

  27. [35]

    Utama, J

    R. Utama, J. Piekarewicz, and H. B. Prosper, Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach, Phys. Rev. C 93, 014311 (2016)

    2016

  28. [36]

    Y. Liu, C. Su, J. Liu, P. Danielewicz, C. Xu, and Z. Ren, Improved naive Bayesian probability classifier in predic- tions of nuclear mass, Phys. Rev. C 104, 014315 (2021)

    2021

  29. [37]

    Liu, H.-L

    G.-P. Liu, H.-L. Wang, Z.-Z. Zhang, and M.-L. Liu, Model-repair capabilities of tree-based machine-learning algorithms applied to theoretical nuclear mass models, Physical Review C 111, 024306 (2025)

    2025

  30. [38]

    Zhang, H.-L

    Z.-Z. Zhang, H.-L. Wang, H.-Y. Meng, and M.-L. Liu, Uncertainty evaluation and correlation analysis of single- particle energies in phenomenological nuclear mean field: an investigation into propagating uncertainties for in- dependent model parameters, Nucl. Sci. Tech. 32, 16 (2021)

    2021

  31. [39]

    M. Liu, Y. Gao, and N. Wang, Statistical errors in Weizs¨ acker-Skyrme mass model, Chin. Phys. C 41, 114101 (2017)

    2017

  32. [40]

    Dobaczewski, W

    J. Dobaczewski, W. Nazarewicz, and P.-G. Reinhard, Er- ror estimates of theoretical models: a guide, J. Phys. G: Nucl. Part. Phys. 41, 074001 (2014)

    2014

  33. [41]

    Y. Luo, J. Peng, and J. Ma, When causal inference meets deep learning, Nat. Mach. Intell. 2, 426 (2020)

    2020

  34. [42]

    Stoitsov, J

    M. Stoitsov, J. Dobaczewski, W. Nazarewicz, and P. Ring, Axially deformed solution of the Skyrme–Hartree–Fock–Bogolyubov equations using the transformed harmonic oscillator basis. The program HFBTHO (v1.66p), Comput. Phys. Commun 167, 43 (2005)

    2005

  35. [43]

    Erler, N

    J. Erler, N. Birge, M. Kortelainen, W. Nazarewicz, E. Olsen, A. M. Perhac, and M. Stoitsov, The limits of the nuclear landscape, Nature 486, 509 (2012)

    2012

  36. [44]

    Stoitsov, N

    M. Stoitsov, N. Schunck, M. Kortelainen, N. Michel, H. Nam, E. Olsen, J. Sarich, and S. Wild, Axially de- formed solution of the Skyrme-Hartree–Fock–Bogoliubov equations using the transformed harmonic oscillator ba- sis (II) hfbtho v2.00d: A new version of the program, Comput....

    2013

  37. [45]

    Mass Explorer, http://massexplorer.frib.msu.edu

  38. [46]

    Bartel, P

    J. Bartel, P. Quentin, M. Brack, C. Guet, and H.-B. H˚ akansson, Towards a better parametrisation of Skyrme- like effective forces: A critical study of the SkM force, Nucl. Phys. A 386, 79 (1982)

    1982

  39. [47]

    Dobaczewski, H

    J. Dobaczewski, H. Flocard, and J. Treiner, Hartree- Fock-Bogolyubov description of nuclei near the neutron- drip line, Nucl. Phys. A 422, 103 (1984)

    1984

  40. [48]

    Chabanat, P

    E. Chabanat, P. Bonche, P. Haensel, J. Meyer, and R. Schaeffer, A Skyrme parametrization from subnuclear to neutron star densities Part II. Nuclei far from stabili- ties, Nucl. Phys. A 635, 231 (1998)

    1998

  41. [49]

    Kl¨ upfel, P.-G

    P. Kl¨ upfel, P.-G. Reinhard, T. J. B¨ urvenich, and J. A. Maruhn, Variations on a theme by Skyrme: A systematic study of adjustments of model parameters, Phys. Rev. C 79, 034310 (2009)

    2009

  42. [50]

    Kortelainen, T

    M. Kortelainen, T. Lesinski, J. Mor´ e, W. Nazarewicz, J. Sarich, N. Schunck, M. V. Stoitsov, and S. Wild, Nuclear energy density optimization, Phys. Rev. C 82, 024313 (2010)

    2010

  43. [51]

    Kortelainen, J

    M. Kortelainen, J. McDonnell, W. Nazarewicz, P.-G. Reinhard, J. Sarich, N. Schunck, M. V. Stoitsov, and S. M. Wild, Nuclear energy density optimization: Large deformations, Phys. Rev. C 85, 024304 (2012)

    2012

  44. [52]

    M. V. Stoitsov, J. Dobaczewski, W. Nazarewicz, S. Pit- tel, and D. J. Dean, Systematic study of deformed nuclei at the drip lines and beyond, Phys. Rev. C 68, 054312 (2003)

    2003

  45. [53]

    M. V. Stoitsov, W. Nazarewicz, and N. Schunck, Large- Scale Mass Table Calculations, Int. J. Mod. Phys. E- nuclear Physics 18, 816 (2009)

    2009

  46. [54]

    A. V. Dorogush, V. Ershov, and A. Gulin, CatBoost: gra- dient boosting with categorical features support (2018), arXiv:1810.11363 [cs.LG]

    2018 arXiv

  47. [55]

    Y. Ding, Z. Chen, W. Lu, and X. Wang, A Cat- Boost approach with wavelet decomposition to im- prove satellite-derived high-resolution PM2.5 estimates in Beijing-Tianjin-Hebei, Atmos. Environ. 249, 118212 16 (2021)

    2021

  48. [56]

    K. Deng, X. Zhang, Y. Cheng, Z. Zheng, F. Jiang, W. Liu, and J. Peng, A remaining useful life prediction method with long-short term feature processing for air- craft engines, Appl. Soft Comput. 93, 106344 (2020)

    2020

  49. [57]

    M. R. Abbasniya, S. A. Sheikholeslamzadeh, H. Nasiri, and S. Emami, Classification of Breast Tumors Based on Histopathology Images Using Deep Features and Ensem- ble of Gradient Boosting Methods, Comput Electr Eng 103, 108382 (2022)

    2022

  50. [58]

    Nguyen, K

    N.-H. Nguyen, K. T. Tong, S. Lee, A. Karamanli, and T. P. Vo, Prediction compressive strength of cement- based mortar containing metakaolin using explainable Categorical Gradient Boosting model, Eng. Struct. 269, 114768 (2022)

    2022

  51. [59]

    M. R. Mumpower, T. M. Sprouse, A. E. Lovell, and A. T. Mohan, Physically interpretable machine learning for nu- clear masses, Phys. Rev. C 106, L021301 (2022)

    2022

  52. [60]

    A. E. Lovell, A. T. Mohan, T. M. Sprouse, and M. R. Mumpower, Nuclear masses learned from a probabilistic neural network, Phys. Rev. C 106, 014305 (2022)

    2022

  53. [61]

    A. E. S. Green, Nuclear Physics (New York, McGraw- Hill, 1955) p. 250

    1955

  54. [62]

    Sobiczewski, F

    A. Sobiczewski, F. Gareev, and B. Kalinkin, Closed shells for Z ¿ 82 and N ¿ 126 in a diffuse potential well, Physics Letters 22, 500 (1966)

    1966

  55. [63]

    Bender, K

    M. Bender, K. Rutz, P.-G. Reinhard, J. A. Maruhn, and W. Greiner, Shell structure of superheavy nuclei in self- consistent mean-field models, Phys. Rev. C 60, 034304 (1999)

    1999

  56. [64]

    Sobiczewski and K

    A. Sobiczewski and K. Pomorski, Description of structure and properties of superheavy nuclei, Prog. Part. Nucl. Phys. 58, 292 (2007)

    2007

  57. [65]

    Q. Mo, M. Liu, and N. Wang, Systematic study of shell gaps in nuclei, Phys. Rev. C 90, 024320 (2014)

    2014

  58. [66]

    Bergstra, B

    J. Bergstra, B. Komer, C. Eliasmith, D. Yamins, and D. D. Cox, Hyperopt: a Python library for model se- lection and hyperparameter optimization, Comput. Sci. Discov 8, 014008 (2015)

    2015

  59. [67]

    Akiba, S

    T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, Optuna: A Next-generation Hyperparameter Optimiza- tion Framework (2019)

    2019

  60. [68]

    J. H. Friedman, Greedy function approximation: A gra- dient boosting machine., Ann. Stat 29, 1189 (2001)

    2001

  61. [69]

    Lundberg and S.-I

    S. Lundberg and S.-I. Lee, A Unified Approach to In- terpreting Model Predictions (2017), arXiv:1705.07874 [cs.AI]

    2017 arXiv

  62. [70]

    Antwarg, R

    L. Antwarg, R. M. Miller, B. Shapira, and L. Rokach, Explaining anomalies detected by autoencoders using Shapley Additive Explanations, Expert Syst. Appl. 186, 115736 (2021)

    2021

  63. [71]

    H. Zhou, Z. Deng, Y. Xia, and M. Fu, A new sampling method in particle filter based on Pearson correlation co- efficient, NEUROCOMPUTING 216, 208 (2016)

    2016

  64. [72]

    Vega Garc´ ıa and J

    M. Vega Garc´ ıa and J. L. Aznarte, Shapley additive ex- planations for NO2 forecasting, Ecological Informatics 56, 101039 (2020)

    2020

  65. [73]

    Liu and C

    X. Liu and C. Aldrich, Explaining anomalies in coal prox- imity and coal processing data with Shapley and tree- based models, Fuel 335, 126891 (2023)

    2023

  66. [74]

    CatBoost, https://catboost.ai/

  67. [75]

    Sobiczewski and Y

    A. Sobiczewski and Y. A. Litvinov, Accuracy of theo- retical descriptions of nuclear masses, Phys. Rev. C 89, 024311 (2014)

    2014

  68. [76]

    Sobiczewski and Y

    A. Sobiczewski and Y. A. Litvinov, Predictive power of nuclear-mass models, Phys. Rev. C 90, 017302 (2014)

    2014

  69. [77]

    Sobiczewski, Y

    A. Sobiczewski, Y. Litvinov, and M. Palczewski, Detailed illustration of the accuracy of currently used nuclear- mass models, At. Data Nucl. Data Tables 119, 1 (2018)

    2018

  70. [78]

    Silwal, C

    R. Silwal, C. Andreoiu, B. Ashrafkhani, J. Bergmann, T. Brunner, J. Cardona, K. Dietrich, E. Dunling, G. Gwinner, Z. Hockenbery, J. Holt, C. Izzo, A. Jacobs, A. Javaji, B. Kootte, Y. Lan, D. Lunney, E. Lykiar- dopoulou, T. Miyagi, M. Mougeot, I. Mukul, T. Murb¨ ock, W. Porter,...

    2022

  71. [79]

    Giraud, L

    S. Giraud, L. Canete, B. Bastin, A. Kankainen, A. Fantina, F. Gulminelli, P. Ascher, T. Eronen, V. Girard-Alcindor, A. Jokinen, A. Khanam, I. Moore, D. Nesterenko, F. de Oliveira Santos, H. Penttil¨ a, C. Petrone, I. Pohjalainen, A. De Roubin, V. Rubchenya, M. Vilen, and J. ¨A...

    2022

  72. [80]

    S. F. Paul, J. Bergmann, J. D. Cardona, K. A. Diet- rich, E. Dunling, Z. Hockenbery, C. Hornung, C. Izzo, A. Jacobs, A. Javaji, B. Kootte, Y. Lan, E. Leistenschnei- der, E. M. Lykiardopoulou, I. Mukul, T. Murb¨ ock, W. S. Porter, R. Silwal, M. B. Smith, J. Ringuette, T. Brun- ...

    2021

  73. [81]

    W. Xian, S. Chen, S. Nikas, M. Rosenbusch, M. Wada, H. Ishiyama, D. Hou, S. Iimura, S. Nishimura, P. Schury, A. Takamine, S. Yan, F. Browne, P. Doornenbal, F. Flav- igny, Y. Hirayama, Y. Ito, S. Kimura, T. M. Kojima, J. Lee, J. Liu, H. Miyatake, S. Michimasa, J. Y. Moon, S. Na...

    2024

  74. [82]

    Mukul, C

    I. Mukul, C. Andreoiu, J. Bergmann, M. Brodeur, T. Brunner, K. A. Dietrich, T. Dickel, I. Dillmann, E. Dunling, D. Fusco, G. Gwinner, C. Izzo, A. Jacobs, B. Kootte, Y. Lan, E. Leistenschneider, E. M. Lykiar- dopoulou, S. F. Paul, M. P. Reiter, J. L. Tracy, J. Dilling, and A. A...

    2021

  75. [83]

    Hukkanen, W

    M. Hukkanen, W. Ryssens, P. Ascher, M. Bender, T. Ero- nen, S. Gr´ evy, A. Kankainen, M. Stryjczyk, O. Be- liuskina, Z. Ge, S. Geldhof, M. Gerbaux, W. Gins, A. Husson, D. A. Nesterenko, A. Raggio, M. Reponen, S. Rinta-Antila, J. Romero, A. de Roubin, V. Virtanen, and A. Zadvor...

    2024 arXiv

  76. [84]

    Z. Ge, M. Reponen, T. Eronen, B. S. Hu, M. Korte- lainen, A. Kankainen, I. D. Moore, D. A. Nesterenko, 17 C. X. Yuan, O. Beliuskina, L. Ca˜ nete, R. de Groote, C. Delafosse, P. Delahaye, T. Dickel, A. de Roubin, S. Geldhof, W. Gins, J. D. Holt, M. Hukkanen, A. Jaries, A. Jokin...

    2024 arXiv

  77. [85]

    Y. M. Xing, C. X. Yuan, M. Wang, Y. H. Zhang, X. H. Zhou, Y. A. Litvinov, K. Blaum, H. S. Xu, T. Bao, R. J. Chen, C. Y. Fu, B. S. Gao, W. W. Ge, J. J. He, W. J. Huang, T. Liao, J. G. Li, H. F. Li, S. Litvinov, S. Naimi, P. Shuai, M. Z. Sun, Q. Wang, X. Xu, F. R. Xu, T. Ya- mag...

    2023

  78. [86]

    Orford, N

    R. Orford, N. Vassh, J. A. Clark, G. C. McLaugh- lin, M. R. Mumpower, D. Ray, G. Savard, R. Surman, F. Buchinger, D. P. Burdette, M. T. Burkey, D. A. Gorelov, J. W. Klimes, W. S. Porter, K. S. Sharma, A. A. Valverde, L. Varriano, and X. L. Yan, Searching for the origin of the ...

    2022

  79. [87]

    Kimura, M

    S. Kimura, M. Wada, H. Haba, S. Ishizawa, Y. Ito, T. Ni- wase, M. Rosenbusch, P. Schury, and A. Takamine, Com- prehensive mass measurement study of 252Cf fission frag- ments with MRTOF-MS and masses of neutron-rich Ce isotopes (2024), arXiv:2402.03873 [nucl-ex]

    2024 arXiv

  80. [88]

    S. Beck, B. Kootte, I. Dedes, T. Dickel, A. A. Kwiatkowski, E. M. Lykiardopoulou, W. R. Plaß, M. P. Reiter, C. Andreoiu, J. Bergmann, T. Brunner, D. Curien, J. Dilling, J. Dudek, E. Dunling, J. Flow- erdew, A. Gaamouci, L. Graham, G. Gwinner, A. Jacobs, R. Klawitter, Y. Lan, E...

    2021

  81. [89]

    X. H. Wu and P. W. Zhao, Predicting nuclear masses with the kernel ridge regression, Phys. Rev. C 101, 051301 (2020)

    2020

  82. [90]

    Wu and C

    X. Wu and C. Pan, Nuclear mass predictions with anisotropic kernel ridge regression, Phys. Rev. C 110, 034322 (2024)

    2024

  83. [91]

    Heenen, J

    P.-H. Heenen, J. Skalski, A. Staszczak, and D. Vretenar, Shapes and α-and β-decays of superheavy nuclei, Nucl. Phys. A 944, 415 (2015)

    2015

  84. [92]

    Jodon, M

    R. Jodon, M. Bender, K. Bennaceur, and J. Meyer, Con- straining the surface properties of effective Skyrme inter- actions, Phys. Rev. C 94, 024335 (2016)

    2016

  85. [93]

    Dutra, O

    M. Dutra, O. Louren¸ co, J. S´ a Martins, A. Delfino, J. R. Stone, and P. Stevenson, Skyrme interaction and nuclear matter constraints, Phys. Rev. C 85, 035201 (2012)

    2012

  86. [94]

    N. Wang, M. Liu, and X. Wu, Modification of nuclear mass formula by considering isospin effects, Phys. Rev. C 81, 044322 (2010)

    2010

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.