Pith. sign in

REVIEW 4 major objections 3 minor 2 cited by

Machine Learning Estimation on the Trace of Inverse Dirac Operator using the Gradient Boosting Decision Tree Regression

T0 review · 4 major / 3 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Given enough labeled data, a bias-corrected gradient boosting estimator reproduces conjugate-gradient trace results for $\text{Tr}\,M^{-n}$ within $1\sigma$, and it works best at a first-order phase transition.

desk verdict Useful, honest feasibility study, but the central claim is only tested for a hybrid estimator that includes up to half exact CG data; P1 results and a proper variance budget are missing. read the letter →

arxiv 2411.18170 v2 pith:FYYJUOJX submitted 2024-11-27 hep-lat

Benjamin J. Choi , Hiroshi Ohno , Takayuki Sumimoto , Akio Tomiya This is my paper · ORCID
classification hep-lat PACS 12.38.Gc11.15.Ha
keywords latticeQCDtraceofinverseDiracoperatorgradientboostingdecisiontreebiascorrectionstochasticestimatorconjugatesolverfirst-orderphasetransitionWilsoncloveraction
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

Lattice QCD calculations often need the trace of an inverse power of the Dirac operator, $\text{Tr}\,M^{-n}$, which normally requires many conjugate-gradient solves. This paper tries to replace most of those solves with a gradient boosting decision tree regressor that learns $\text{Tr}\,M^{-n}$ from cheaper observables such as the plaquette, the Polyakov loop, or lower inverse powers of the Dirac operator. The proposed estimator applies a bias correction: the average residual on a labeled bias-correction set is added to the machine-learning predictions on the unlabeled set. The authors report that with a large enough labeled fraction this corrected estimator matches the original conjugate-gradient trace within $1\sigma$ and with statistical error no more than 1.2 times the original, with the smallest labeled fraction ($R_{\rm LB}\gtrsim10\%$, the fraction of configurations with known CG traces) needed on the dataset containing a first-order phase transition. If this holds, the expensive inversion would only be needed on a small labeled subset of configurations, and the rest of the ensemble would be predicted essentially for free.

What carries the argument

The load-bearing mechanism is the gradient boosting decision tree regressor combined with the bias-correction prescription of Eqs. (1)--(2). Gradient boosting builds $f(X)$ by adding shallow decision trees sequentially, each new tree fitting the residual left by the previous ensemble; the implementation used here runs 40 depth-3 trees with learning rate 0.1 and sub-sampling 0.7. The bias correction is what lets the model be wrong configuration-by-configuration as long as it is right on average: $\bar{Y}^{P1}$ transfers the mean residual computed on the labeled bias-correction set $S_{\rm BC}$ to the mean prediction on the unlabeled set $S_{\rm UL}$, and $\bar{Y}^{P2}$ blends this with the direct CG average on all labeled data. The acceptance criteria are EC-1 (the two means agree within each other's $1\sigma$ error, called score 2) and EC-2 ($R_\sigma=\sigma_Q/\sigma_{\rm Orig}\le1.2$), with errors estimated by $10{,}000$ bootstrap resamples. The method's precondition, checked in the paper's correlation matrices, is a strong correlation between the chosen input observable and the target trace.

What would settle it

A direct test is to rerun the analysis with the same $X,Y$ pairs and the same claimed-good values such as $R_{\rm LB}=30\%$, $R_{\rm TR}=40\%$ for $(X,Y)=({\rm Plaquette},\text{Tr}\,M^{-3})$ on the ID-0 data, but with the CG traces already known for the unlabeled set, and then compare $\bar{Y}^{P1}$ against the true mean on $S_{\rm UL}$. The assumption fails if the difference is significantly larger than the mean residual on $S_{\rm BC}$.

Watch

Extended reading notes

Core claim

The central claim is that the bias-corrected estimators of Eqs. (1)--(2), obtained with the gradient boosting decision tree regression, reproduce the original conjugate-gradient (CG) trace estimates for $\text{Tr}\,M^{-n}$ within $1\sigma$ and with $R_\sigma \le 1.2$ in well-defined regions of the two resolution parameters: the labeled fraction $R_{\rm LB}=N_{\rm LB}/N$ and the training fraction $R_{\rm TR}=N_{\rm TR}/N_{\rm LB}$. The first estimator $\bar{Y}^{P1}$ adds the mean residual on the bias-correction set $S^{Y}_{\rm BC}$ to the mean machine-learning prediction on the unlabeled set $S^{Y}_{\rm UL}$; the second estimator $\bar{Y}^{P2}$ takes the weighted average of $\bar{Y}^{P1}$ with the direct CG mean on the labeled set. On the heaviest-quark (ID-0) data the method succeeds for $n=1,2,3$ at $R_{\rm LB}\gtrsim40\%$, with examples such as $X={\rm Plaquette}$, $Y=\text{Tr}\,M^{-3}$ succeeding already at $R_{\rm LB}\gtrsim30\%$, $R_{\rm TR}\lesssim40\%$; on the first-order-transition (ID-1) data good agreement starts at $R_{\rm LB}\gtrsim10\%$ for $n=1,2,3$; on the lightest-quark (ID-2) data only $n=1,2$ work, at $R_{\rm LB}\gtrsim40\%$. The trace $\text{Tr}\,M^{-4}$ is harder in every dataset, requiring $R_{\rm LB}\gtrsim40\%$ even at the phase transition and working best with input $X=\text{Tr}\,M^{-3}$. The paper concludes that ML estimation of $\text{Tr}\,M^{-n}$ works better with heavier quark mass and works 'quite well' when the dataset contains the first-order phase transition.

Load-bearing premise

The average model error on the labeled bias-correction set is assumed to equal the error on the unlabeled set, an assumption the paper itself flags as potentially suffering from 'unwanted fluctuations or weird bias due to the partial data usage'.

Editorial extensions

If this is right

  • If the central claim holds, trace estimation can be reduced to labeling a fraction $R_{\rm LB}$ of configurations with the CG solver, training the regressor, and predicting the rest, cutting the dominant cost of the calculation.
  • The method performs best on the dataset containing a first-order phase transition, where $R_{\rm LB}\gtrsim10\%$ suffices for $n=1,2,3$, so the cheap estimator is most useful precisely where the trace signal is hardest to obtain.
  • For $\text{Tr}\,M^{-4}$ the labeled fraction has to stay near $R_{\rm LB}\gtrsim40\%$ in all three datasets, and $X=\text{Tr}\,M^{-3}$ is the recommended input.
  • The paper's EC-1 score-2 plus $R_\sigma\le1.2$ rule gives a practical, tentative criterion for deciding when the ML estimator can replace the CG result.
  • The planned use of the method is to compute chiral order-parameter cumulants such as susceptibility, skewness, and kurtosis from the ML traces, which the paper lists as work in progress.

Reading between the lines

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

  • The paper does not test whether the bias correction transfers across ensembles; a natural extension is to train on one $\kappa$ value and bias-correct on another, which would reveal whether the good regions survive outside the single-ensemble setting.
  • Because Eq. (1) applies a single additive bias correction, any input-dependent error (for example, larger error in one phase than the other) survives; fitting a second regressor to the residuals and re-checking the good regions would test how much this matters.
  • The paper does not report wall-clock timings, so the claimed cost reduction is implicit; measuring CG solver time against training and prediction time would show whether the $R_{\rm LB}=10\%$--$40\%$ regions actually save time in production.
  • The strong performance at the first-order phase transition suggests the model is learning the two-phase mixture; checking whether the predicted traces on $S_{\rm UL}$ reproduce the bimodal distribution of the CG data would make that mechanism visible.
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 / 3 minor

Summary. The manuscript reports a preliminary machine-learning study of the trace of inverse Dirac operator, Tr M^{-n}, in 4-flavor Wilson-clover QCD on three 16^3 x 4 ensembles with different hopping parameters. Using LightGBM gradient-boosted decision trees, the authors train a model f(X) to map input observables X (Plaquette, Polyakov loop, or Tr M^{-m} with m<n) to Y = Tr M^{-n}. They construct a bias-corrected ML estimator \bar Y_P1 (Eq. 1) and a hybrid estimator \bar Y_P2 (Eq. 2) that averages \bar Y_P1 with the exact CG estimate on the labeled subset. The paper defines two evaluation criteria: EC-1 (1-sigma agreement of means) and EC-2 (statistical error ratio R_sigma <= 1.2 relative to the original CG result). For each (X,Y) pair and each dataset, they report regions in the labeled-fraction R_LB and training-fraction R_TR plane where P2 achieves EC-1 score 2 and EC-2. The reported good regions are summarized in Tables 3-5, with the main qualitative conclusions that heavier quark mass and the first-order phase-transition ensemble (ID-1) give larger good regions, while Tr M^{-4} is generally harder.

Significance. If the claimed results held, the paper would demonstrate a useful cost-reduction strategy for stochastic trace estimation in lattice QCD, building on the bias-correction methodology of Yoon et al. The authors are honest about the preliminary nature of the work, explicitly list to-do items, fix the LightGBM hyperparameters, and provide correlation matrices that motivate the choice of inputs. However, the central claim is currently supported only for the hybrid estimator \bar Y_P2, not for the pure machine-learning estimator \bar Y_P1. Since \bar Y_P2 contains an exact-CG component with weight N_LB/N (30-50% in the reported good regions), the agreement with the full CG result is partly built in by construction. The paper also does not account for the training-set split variance in the bootstrap errors and uses a post-hoc EC-2 threshold. These issues make the significance of the quantitative results uncertain, although the question is worthy of investigation and the manuscript provides a reasonable framework for a subsequent, more complete analysis.

major comments (4)
  1. [§2, Eq. (1)] The central claim that gradient boosting estimates Tr M^{-n} is evaluated only through the hybrid estimator \bar Y_P2, not through the machine-learning estimator \bar Y_P1 of Eq. (1). Since \bar Y_P2 = (N_UL/N)\bar Y_P1 + (N_LB/N)\bar Y^{(LB)}, at the reported good regions R_LB is 30-50% (and 10% in Table 5), so a substantial fraction of \bar Y_P2 is the exact CG average over the labeled subset. Because \bar Y^{(LB)} estimates the same ensemble average as \bar Y^{Orig} from a subset of the same configurations, EC-1 score 2 and R_sigma <= 1.2 can be satisfied even if \bar Y_P1 is badly biased or noisy. No results for \bar Y_P1 are reported anywhere, and no control with R_TR = 0 is shown as a baseline. Tables 3-5 therefore establish only that a weighted average containing an exact-CG component agrees with the full CG result; the title and the summary claim about ML estimation outrun this evidence. Please report \bar Y_P1 (and its EC scores) for the same grid, or restrict the claims to \bar Y_P2.
  2. [§2, Eq. (1)] The bias-correction step adds the mean residual on S_BC to the ML predictions on S_UL. This assumes that the model bias on the unlabeled configurations equals the bias on the labeled configurations. The paper itself cautions about "unwanted fluctuations or weird bias due to the partial data usage" (Section 2), yet no diagnostic is given: for example, one could hold out a second labeled slice and compare \bar Y_P1 computed on it with the direct CG average, or examine residual distributions across bootstrap resamples. Without such a check, unbiasedness of \bar Y_P1 is an assumption, and by extension the EC-1 agreements and the good regions in Tables 3-5 do not demonstrate that the ML estimate matches CG on unlabeled data.
  3. [§2, bootstrap error] The bootstrap error for \bar Y_P1 and \bar Y_P2 is computed by resampling configurations with N_BS = 10,000, but the text does not state that the model is retrained on each bootstrap sample. With a fixed trained model, the bootstrap variance omits the training-set split variance, i.e., the randomness in choosing S_TR and S_BC. This makes sigma_Q too small and likely biases R_sigma below its true value. Please clarify whether retraining is performed inside the bootstrap; if not, use nested resampling or a separate Monte Carlo over splits, and report both variance components.
  4. [§3, EC-2] The success threshold R_sigma <= 1.2 is described as "tentatively determined empirically, monitoring our preliminary results" and is applied to the same datasets that were monitored. This makes the criterion post-hoc; the reported good regions may reflect selection of the threshold rather than a robust property of the estimator. Please provide a pre-specified or independently motivated threshold, or show that the good regions are insensitive to the threshold within a reasonable range (e.g., 1.1-1.5).
minor comments (3)
  1. [Table 1] The caption says the ID-1 row is bold, but in the manuscript text the row does not appear bold; please fix the formatting or the caption.
  2. [Figs. 3-5] The color scale for R_sigma in the EC-2 panels is not numerically labeled, so the reader cannot read off the actual R_sigma values; adding a quantitative color bar would improve reproducibility.
  3. [§2, hyperparameters] The LightGBM hyperparameters (40 stages, depth 3, learning rate 0.1, subsampling 0.7) are fixed, but no validation or sensitivity study is reported; a sentence on how these values were selected would help assess robustness.

Circularity Check

1 steps flagged · score 6.0 of 10

Good regions are reported only for the hybrid estimator P2, which injects the exact CG trace of the labeled subset directly into the 'ML estimate'; the pure ML estimator P1 is never validated against CG, so the headline claim is partially forced by construction.

  1. self definitional [Sec. 2, Eq. (2); Sec. 3, Tables 3-5 and Figs. 3-5]
    "¯𝑌P2 = 𝑁UL/𝑁 ¯𝑌P1+ 𝑁LB/𝑁 ¯𝑌(LB) where 𝑁 =𝑁LB+𝑁UL, ¯𝑌(LB) = 1/𝑁LB ∑︁ 𝑌𝑘∈𝑆𝑌 LB 𝑌𝑘. (2) ... Table 3: Results on {R≳ LB,R ≲ TR} of P2 estimation for ID-0 dataset where the heaviest quark is used."

    The estimator reported in all success tables is P2, which is defined as a weighted average of the ML estimate P1 and the exact CG trace on the labeled subset, with weight N_LB/N. In the claimed good regions R_LB=30-50%, so 30-50% of P2 is exact target data that enters without any ML model. A pure average of the labeled subset alone would already agree with the full CG average at the ~0.5-1 sigma level, making EC-1 score-2 and R_sigma<=1.2 easy to satisfy even if P1 were useless. Since Tables 3-5 and Figs. 3-5 report only P2 and no comparison of P1 with the CG result is shown, the paper's conclusion that the GBDT ML estimation of Tr M^{-n} 'works well' is not actually tested independently of the exact CG input baked into P2.

full rationale

The derivation chain is not circular at the level of Eq. (1): the bias-corrected ML estimator P1 is a standard supervised fit, and the GBDT training on S_TR is genuine. Ref. [8] is used only as a data source, not as a load-bearing argument. However, the central validation step is partially self-referential: the hybrid estimator P2 (Eq. 2) contains the exact CG average of the labeled subset with weight N_LB/N, and all reported good regions (Tables 3-5, Figs. 3-5) are for P2 only. At R_LB=30-50%, this exact component alone makes agreement with the full CG estimate statistically likely, so the success criterion does not establish that the ML model reproduces Tr M^{-n}. The R_sigma<=1.2 threshold is also 'tentatively determined empirically, monitoring our preliminary results,' so it is calibrated on the same data used for validation; that is an evaluation-standard weakness but not an equation-level reduction. These issues are partial rather than total circularity: a proper P1-only validation or a control subtracting the labeled component could have supported the claim.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the supervised-learning setup: a strong input-output correlation is assumed sufficient, the bias measured on a labeled subset is assumed to transfer to the unlabeled set, and the reported bootstrap errors assume a fixed trained model. The paper introduces no new entities; the hand-set numbers are the LightGBM hyperparameters and the post-hoc EC-2 threshold R_sigma <= 1.2.

free parameters (2)
  • LightGBM hyperparameters = 40 boosting stages, depth 3, learning rate 0.1, sub-sampling 0.7
    Set without a tuning procedure in Section 2; the resulting score grids depend on these choices, and no sensitivity study is provided.
  • EC-2 threshold R_sigma = 1.2
    Declared 'tentatively determined empirically, monitoring our preliminary results' (Section 2); the success criterion is calibrated on the same datasets used to draw the conclusions.
assumptions (4)
  • domain assumption Strong correlation between input X and output Y is sufficient for the gradient boosting regression to learn an accurate mapping.
    Section 2 states 'the strong correlation between X and Y is often required' and Fig. 1 is used to justify the approach; learnability from correlation is assumed, not proven.
  • domain assumption The average bias measured on S_BC equals the average bias on S_UL.
    Eq. (1) transfers the residual from the bias-correction set to the unlabeled set; Section 2 warns of possible 'unwanted fluctuations or weird bias', but no distributional argument is given.
  • domain assumption Bootstrap resampling over configurations with a fixed trained model yields valid errors for the ML estimator.
    Section 2 describes bootstrap with N_BS=10,000, but the model is trained once on S_TR; the training-set sampling variance is not propagated into sigma_Q.
  • domain assumption The CG trace labels are treated as exact values in training, with their statistical errors ignored in the loss.
    Standard practice in ML regression on lattice data; the paper does not account for label noise from the stochastic-source CG estimates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Machine Learning Estimation on the Trace of Inverse Dirac Operator using the Gradient Boosting Decision Tree Regression." pith.science (2026). https://pith.science/paper/FYYJUOJX

@misc{pith2026241118170,
  author       = {Pith},
  title        = {Pith review of: Machine Learning Estimation on the Trace of Inverse Dirac Operator using the Gradient Boosting Decision Tree Regression},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FYYJUOJX}},
  note         = {Machine review of arXiv:2411.18170}
}
abstract

We present our preliminary results on the machine learning estimation of $\text{Tr} \, M^{-n}$ from other observables with the gradient boosting decision tree regression, where $M$ is the Dirac operator. Ordinarily, $\text{Tr} \, M^{-n}$ is obtained by linear CG solver for stochastic sources which needs considerable computational cost. Hence, we explore the possibility of cost reduction on the trace estimation by the adoption of gradient boosting decision tree algorithm. We also discuss effects of bias and its correction.

Figures

Figures reproduced from arXiv: 2411.18170 by the authors.

Figure 1
Figure 1. Correlation between physical observables. Here, we also monitor RTR = 0% (𝑆 𝑍 TR = ∅ for 𝑍 = 𝑋, 𝑌) to check that 𝑆 𝑌 LB itself approaches to the true answer along the increase of RLB, that is, we do not perform ML estimation at RTR = 0% but observe only the statistical average and error of the 𝑆 𝑌 LB. We also monitor RTR = 100% (𝑆 𝑍 BC = ∅) to check what happens when we do not use the bias correction method in the M… view at source ↗
Figure 2
Figure 2. Histograms on the three possible cases can be occurred where blue (red) histogram represents original CG result (ML estimation). Here, data for 𝑋 = Tr 𝑀−1 and 𝑌 = Tr 𝑀−2 are used. Vertical solid (dotted) lines represent statistical averages (1 𝜎 errors). 4 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Results on (a) EC-1 and (b) EC-2 of P2 estimation using (𝑋 , 𝑌) = (Plaquette , Tr 𝑀−3 ) for ID-0 dataset where the heaviest quark is used. White, orange, red color of (a) represents score 2, 1, 0, respectively. Here, the magnitude of R𝜎 (Eq. (3)) in (b) is represented using the color of rainbow. For example, if R𝜎 ≈ 7 for certain point, then we observe purple-like color there.  R ≳ LB, R ≲ TR [PITH_FULL_IMAGE:figu… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Results on (a) EC-1 and (b) EC-2 of P2 estimation using (𝑋 , 𝑌) = (Plaquette , Tr 𝑀−3 ) for ID-2 dataset where the lightest quark is used. We use same notation as in [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Results on (a) EC-1 and (b) EC-2 of P2 estimation using (𝑋 , 𝑌) = (Plaquette , Tr 𝑀−3 ) for ID-1 dataset where the first order phase transition observed. We use same notation as in [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Machine Learning-Based Estimation of Cumulants of Chiral Condensate via Multi-Ensemble Reweighting with Deborah.jl

    hep-lat 2026-02 conditional novelty 5.0 of 10

    Using Tr M^-1 as both an input and a feature, a bias-corrected ML model predicts Tr M^-2..-4 and reproduces chiral-condensate cumulants with ~1% labeled data at ~26% of the original cost.

  2. Investigating the axial structure of the nucleon based on large-volume lattice QCD at the physical point

    hep-lat 2025-05 conditional novelty 5.0 of 10

    Using large-volume lattice QCD at physical quark masses, the PACS Collaboration shows that a new excited-state subtraction makes the nucleon pseudoscalar and induced-pseudoscalar form factors consistent with pion-pole...

Reference graph

Works this paper leans on

15 extracted references · 9 canonical work pages · cited by 2 Pith papers

  1. [1]

    S.-J.DongandK.-F.Liu, StochasticestimationwithZ(2)noise , Phys.Lett.B 328(1994)130 [hep-lat/9308015]

    arXiv 1994

  2. [2]

    Duane and J.B

    S. Duane and J.B. Kogut,Hybrid Stochastic Differential Equations Applied to Quantum Chromodynamics, Phys. Rev. Lett.55(1985) 2774

    work page 1985

  3. [3]

    Duane and J.B

    S. Duane and J.B. Kogut,The Theory of Hybrid Stochastic Algorithms, Nucl. Phys. B275 (1986) 398

    work page 1986

  4. [4]

    Duane, A.D

    S. Duane, A.D. Kennedy, B.J. Pendleton and D. Roweth,Hybrid Monte Carlo, Phys. Lett. B 195 (1987) 216

    work page 1987

  5. [5]

    Friedman,Greedy function approximation: A gradient boosting machine., The Annals of Statistics29(2001) 1189

    J.H. Friedman,Greedy function approximation: A gradient boosting machine., The Annals of Statistics29(2001) 1189

    work page 2001

  6. [6]

    B. Yoon, T. Bhattacharya and R. Gupta,Machine Learning Estimators for Lattice QCD Observables, Phys. Rev. D100 (2019) 014504 [1807.05971]

    arXiv 2019

  7. [7]

    J. Kim, G. Pederiva and A. Shindler,Machine learning mapping of lattice correlated data, Phys. Lett. B856 (2024) 138894 [2402.07450]. 8 Trace estimation with gradient boosting decision tree regression Benjamin J. Choi

    arXiv 2024

  8. [8]

    H. Ohno, Y. Kuramashi, Y. Nakamura and S. Takeda,Continuum extrapolation of the critical endpoint in 4-flavor QCD with Wilson-Clover fermions, PoS LATTICE2018(2018) 174 [1812.01318]

    arXiv 2018

Show all 15 references
  1. [9]

    Boku, K.-I

    T. Boku, K.-I. Ishikawa, Y. Kuramashi and L. Meadows,Mixed Precision Solver Scalable to 16000 MPI Processes for Lattice Quantum Chromodynamics Simulations on the Oakforest-PACS System, in5th International Workshop on Legacy HPC Application Migration: InternationalSymposiumonCo...

    2017 arXiv

  2. [10]

    Nakamura and H

    Y. Nakamura and H. Stuben,BQCD - Berlin quantum chromodynamics program, PoS LATTICE2010(2010) 040 [1011.0199]

    2010 arXiv

  3. [11]

    Sheikholeslami and R

    B. Sheikholeslami and R. Wohlert,Improved Continuum Limit Lattice Action for QCD with Wilson Fermions, Nucl. Phys. B259 (1985) 572

    1985

  4. [12]

    Iwasaki,Renormalization group analysis of lattice theories and improved lattice action: Two-dimensional non-linear O(N) sigma model, Nucl

    Y. Iwasaki,Renormalization group analysis of lattice theories and improved lattice action: Two-dimensional non-linear O(N) sigma model, Nucl. Phys. B258 (1985) 141

    1985

  5. [13]

    Iwasaki,Renormalization Group Analysis of Lattice Theories and Improved Lattice Action

    Y. Iwasaki,Renormalization Group Analysis of Lattice Theories and Improved Lattice Action. II. Four-dimensional non-Abelian SU(N) gauge model, 1111.7054

    arXiv

  6. [14]

    G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Maet al.,LightGBM: A highly efficient gradient boosting decision tree, Advances in Neural Information Processing Systems30 (2017)

    2017

  7. [15]

    Blaom, F

    A.D. Blaom, F. Kiraly, T. Lienart, Y. Simillides, D. Arenas and S.J. Vollmer,MLJ: A julia package for composable machine learning, Journal of Open Source Software5(2020) 2704. 9

    2020

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.