Pith. sign in

REVIEW 3 major objections 4 minor 45 references

Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas

T0 review · 3 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a cycle-consistent surrogate—a conditional U-Net forward model paired with an optimization-based inverse—can reproduce the two-dimensional tokamak edge-plasma state from five control parameters with under 2.6 percent

desk verdict Solid forward and inverse surrogates on held-out SOLPS data, but the abstract over-attributes Γcore recovery to cycle consistency; the ablation measures the training objective, not inverse benefit. read the letter →

arxiv 2607.21407 v1 pith:IY5X5SH5 submitted 2026-07-23 physics.comp-ph cs.AIphysics.plasm-ph

Abdourahmane Diaw , Sebastian De Pascuale , Jae-Sun Park , Ivan Paradela Perez , Jeremy D. Lore , Stefan Dasbach This is my paper · ORCID
classification physics.comp-phcs.AIphysics.plasm-ph PACS 52.55.Fa52.65.-y
keywords SOLPS-ITERedgeplasmaneuralsurrogatecycleconsistencyuncertaintyquantificationinversemodelingtokamakU-Net
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 aims to make tokamak edge-plasma modeling fast enough for real-time control, parameter scans, and inverse analysis by replacing multi-hour simulations of the scrape-off layer with a neural surrogate. It claims that a conditional U-Net trained on about 760 converged runs reproduces the four key plasma fields (electron and ion temperature, density, parallel velocity) with normalized errors below 2.6 percent and Pearson correlations above 0.95. The central new idea is cycle consistency: parameters recovered from a field, when pushed back through the forward model, must reproduce that field. This raises a self-consistency score from 0.59 to 0.99 without hurting forward accuracy, and it makes the inverse recover all five control parameters (including the core fueling rate, previously the hardest) with correlation at or above 0.97. A warm-started database built with a branching k-d tree reaches a completion rate above 95 percent, and each forward evaluation takes milliseconds—five to six orders of magnitude faster than the original solver.

What carries the argument

The load-bearing mechanism is the cycle-consistency identity linking the forward map and the inverse: a parameter vector c maps to a field Fθ(c), and the inverse G⋆(Fθ(c)) should return c, while a target field y maps through G⋆(y) to parameters whose forward pass Fθ(G⋆(y)) should approximate y. The forward model is a U-Net (a convolutional network that preserves fine spatial detail via skip connections) conditioned on the five scalars through FiLM layers and restricted to the plasma domain by a binary mask; the optimization-based inverse G⋆ uses Adam through the frozen network, warm-started by a learned convolutional pseudo-inverse Gψ and refined with random restarts. The cycle constraint re

What would settle it

Compute the numerical Jacobian of the trained forward model with respect to the five inputs at a grid of held-out points. If the smallest singular value is near zero relative to the largest, there is a direction in parameter space that the fields do not encode; then two very different parameter vectors produce nearly identical fields, and the claimed r≥0.97 parameter recovery would fail for noisy or out-of-distribution inputs.

Watch

Extended reading notes

Core claim

The paper introduces a three-stage surrogate for edge plasmas: a conditional U-Net forward model maps five scalar control parameters (input power, core fueling, gas puff, and two cross-field transport coefficients) to the two-dimensional fields Te, Ti, ne, and parallel velocity; an optimization-based inverse recovers the parameters from a target field by differentiating through the frozen forward model; and a cycle-consistency stage passes the recovered parameters back through the forward model, giving a label-free quality check. On a held-out test set, all four fields achieve Pearson correlation above 0.95 and NRMSE below 2.6 percent, with parallel velocity the hardest (r=0.975). Cycle-cons

Load-bearing premise

The five control parameters are assumed to determine the two-dimensional steady-state fields uniquely enough that optimizing through the frozen forward model recovers the true parameters; the paper itself notes that the forward mapping is locally insensitive to some parameter combinations and that a clean cycle reconstruction does not by itself prove the parameters are identified.

Editorial extensions

If this is right

  • Parameter scans and sensitivity analysis across the five-dimensional operating space become feasible in milliseconds per evaluation, enabling broad exploration of detachment regimes.
  • The differentiable inverse offers an automated route to infer cross-field transport coefficients and fueling rates from measured or simulated fields, replacing hand-tuned iterative fitting.
  • Real-time plasma control loops (1–10 ms) could embed edge-plasma state estimation, since a forward pass takes about 16 ms on a single CPU core and sub-millisecond on a GPU.
  • The cycle-consistency round-trip error acts as a self-supervised reliability flag: at inference, predictions whose reconstructed fields deviate from the target can be marked untrustworthy without ground-truth parameters.
  • The committee ensemble flags the inner divertor near the strike point as the least certain regime (about 13 percent of points flagged), directing new expensive simulations to where they are most needed.

Reading between the lines

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

  • Beyond the paper: the same cycle-consistency round trip could serve as an out-of-distribution detector for any expensive surrogate—a large cycle error on a new query would signal regime shift even when no ground truth is available.
  • A testable extension is to train a second pseudo-inverse directly on the cycle-error gradient, potentially cutting the 1200-step Adam descent and making real-time inversion practical.
  • Because the inner divertor's near-strike-point electron temperature is bimodal across the database, active-learning acquisition focused there might resolve the observed R2≈0.89 and simultaneously improve inverse identifiability of parameters that set detachment onset.
  • The k-d tree warm-start is effectively numerical continuation in five dimensions; replacing the Euclidean distance with a physics-informed metric (e.g., weighting parameters by local sensitivity in the forward model) could further reduce the residual failure rate.
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.

Referee Report

3 major / 4 minor

Summary. The paper introduces a cycle-consistent neural surrogate for SOLPS-ITER edge-plasma simulations of a DIII-D lower-single-null configuration. A conditional U-Net Fθ maps five control parameters (Ptot, Γcore, ΓD2, D⊥, χi) plus a geometry mask to 2D fields (Te, Ti, ne, ua). A differentiable optimizer through the frozen forward model, warm-started by a learned pseudo-inverse Gψ, recovers parameters from target fields; a cycle-consistency term in joint training regularizes the forward/pseudo-inverse pair. A separate committee of MLPs predicts 1D Te/ne profiles with quality flags. On a 152-run held-out test set, the forward model reports NRMSE below 2.6% and Pearson r above 0.975 for all fields, and inverse recovery reports r ≥ 0.97 for all five parameters at the adopted cycle weight. The paper also proposes a k-d tree warm-start strategy for database generation and claims a completion rate above 95% versus roughly 30% failures in a prior cold-started ensemble.

Significance. If the results hold, this is a valuable contribution to surrogate-based edge-plasma modeling: a fast, differentiable forward model, a practical inverse pipeline for parameter recovery, a self-supervised consistency check, and an uncertainty-aware 1D profile ensemble. The forward-accuracy evaluation is carefully done on a properly split held-out set, the convergence detection in Appendix B is thoughtful, and the code/data availability statements are a real strength. The authors also state important limitations (single geometry, no experimental validation, unenforced conservation laws), which is commendable. The main weakness is that the paper's central causal claim — that cycle-consistency regularization enables recovery of the core fueling rate — is not supported by the experiments as reported, because the ablation metric is essentially the training objective and no λcyc = 0 control is run through the inference-time inverse pipeline.

major comments (3)
  1. [Sec. 4.3 / Eq. (7), Fig. 4, Table 3] The average cyclical R2 reported in the ablation is computed by sweeping control parameters through Fθ and then Gψ and comparing the recovered values to the swept values. This is exactly the quantity minimized by the cycle term in Eq. (7), so increasing λcyc is expected to raise it; it is not a test of the inference-time optimizer G⋆ of Eq. (5). The abstract's claim that cycle-consistency regularization 'enables recovery of Γcore' is therefore unsupported: Sec. 4.4 reports inverse recovery only at λcyc=0.5, with no λcyc=0 checkpoint pushed through the same optimization-based inverse. Please add the missing λcyc=0 (and ideally intermediate) inverse-recovery results for all five parameters, or revise the causal claim. The manuscript's own caveats in Secs. 4.3 and 4.4 — that pair self-consistency need not imply invertibility and that good cycle reconstruction does not guarantee parameter id
  2. [Secs. 3.3 and 4.4] The paper presents Lcycle (Eq. 6) as a self-supervised quality check that 'requires no ground-truth labels at inference,' but it never validates that this metric is actually informative about parameter-recovery error. Given the explicit statement in Sec. 4.4 that good cycle reconstruction does not guarantee accurate recovery, the utility of Lcycle as a reliability diagnostic is unproven. Please provide direct evidence — e.g., a scatter or reliability diagram of Lcycle versus ∥ĉ − ctrue∥ on held-out cases — or temper the claim that the cycle loss functions as a quality check.
  3. [Secs. 2 and 6] The k-d tree warm-start completion-rate improvement (>95% versus ~70%) is supported only by a comparison to a different cold-started ensemble from Ref. [22], which used different runs and settings. This is not a same-design controlled comparison, so the quantitative gain attributable to the warm-start strategy is not established by the data reported here. Either run a cold-start control on the same LHS design and convergence criterion, or revise the claims to state only that the k-d tree strategy achieved >95% completion, without attributing a specific improvement relative to Ref. [22].
minor comments (4)
  1. [Eq. (10)] The quality score is written as 'si = k σi S', but the text and threshold 'si ≥ 1' imply si = k σi / S. Please fix the notation and define S immediately before use.
  2. [Table 3 / Fig. 4] The ablation reports point estimates of test MSE and cyclical R2 without repeated-seed variation, even though the text notes run-to-run variation when training from scratch. Reporting mean ± std over a few seeds for each λcyc would strengthen the comparison.
  3. [Appendix B] The window lengths Ttail and L are defined in terms of N_tail and N_win, but the numerical values of these multipliers are never stated. Please give the actual window sizes used in the convergence criterion.
  4. [Ref. [8]] Reference [8] is a disruption-prediction paper; it does not obviously support the statement about existing edge-plasma surrogates. Consider replacing it with a more direct citation to edge-plasma surrogate work.

Circularity Check

1 steps flagged · score 4.0 of 10

Cyclical R2 metric is the training objective in disguise; held-out inverse recovery is independent but the causal role of cycle consistency is uncontrolled.

  1. self definitional [Sec. 3.4 / Eq. (7); Sec. 4.3 / Fig. 4 / Table 3]
    "min θ,ψ ρ(Fθ(c,m),y)+ρ(Gψ(y,m),c)+λcyc||c−Gψ(Fθ(c,m),m)|| ... for each control parameter, we sweep it linearly across its range ... and measure the coefficient of determination between the swept and recovered values ... the average cyclical R2 increases from 0.59 at λcyc=0 to 0.99 at λcyc=0.5"

    The reported 'average cyclical R2' is computed from pairs (c, Gψ(Fθ(c,m))) — exactly the cycle term minimized in Eq. (7). Increasing λcyc directly weights this loss, so the observed rise 0.59→0.99 is a restatement of the training objective rather than an independent measure of forward-model invertibility. The paper concedes this: 'it does not by itself imply that the forward model has become more invertible.' The abstract nevertheless credits cycle consistency with 'enabl[ing] recovery of the core fueling rate'; Sec. 4.4's true parameter recovery is an independent held-out test, but no λcyc=0 control is run through the same inverse pipeline, so the causal attribution is not established.

full rationale

The paper's principal forward and inverse claims are tested against held-out SOLPS-ITER runs and true parameters, so the core derivation is not circular. The forward U-Net NRMSE/Pearson numbers, the k-d tree completion rate, and the inverse parameter recovery (Sec. 4.4, Fig. 5) are all evaluated against independent data. The one self-referential step is the cyclical R2 used as evidence of 'invertibility gains': the metric is defined as the cycle loss minimized in Eq. (7), and the reported improvement with λcyc is expected by construction. The paper itself flags that this does not imply invertibility, and Sec. 4.4 warns that good cycle reconstruction does not guarantee parameter identification. Because the central inverse-recovery result does not depend on the cycle metric and is validated against true held-out parameters, the circularity is partial and confined to a supporting ablation. The absence of a λcyc=0 comparison in the inverse pipeline is an experimental gap rather than a circular reduction, so it is noted in the step but does not by itself raise the score above 4.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

No new physical entities are postulated; the learned pseudo-inverse Gψ is a network component, not a physical object. The central claims rest on the axioms above: simulation-as-truth, identifiability of the 5-parameter description, split validity, UQ proxy validity, and the external-baseline comparison for warm-start gains. The free parameters listed are the hand-selected or test-selected hyperparameters that materially affect the reported numbers.

free parameters (7)
  • λcyc (cycle-consistency weight) = 0.5
    Chosen by sweeping λcyc and selecting the value with best held-out test MSE/cyclical R2; drives all cycle-consistency claims.
  • λg (gradient loss weight) = 0.2
    Hand-set in composite loss Eq. 8; affects sharp-feature fidelity and gradient penalty.
  • Per-channel loss weights wc = Te/Ti:1.0, ne:1.2, ua:1.5
    Hand-set priorities in base loss; influences the reported error balance across output fields.
  • Inverse L2 regularization coefficient λreg = not reported
    Used in Eq. 5 to prevent extreme parameter drift; value not given, so inverse results cannot be exactly reproduced without code inspection.
  • Convergence thresholds εrel, zmax = 0.01, 2
    Define the steady-state acceptance rule that determines which runs enter the database; changing them changes all downstream metrics.
  • Quality-flag threshold k and scaling S = k=2; S=σ̄Ēcal/Ēens
    Determines which points are flagged for further simulation; sets the reported 13%/3% inner/outer target flag fractions.
  • Random-restart noise σ_noise and count Nr = 0.2, 5
    Hyperparameters of the inverse optimization; affect recovery robustness and computational cost.
assumptions (6)
  • domain assumption SOLPS-ITER (B2.5-EIRENE) output is ground truth for training and evaluation.
    Every accuracy claim compares against these simulations; the paper states the surrogate has not yet been validated against experimental DIII-D measurements (Sec. 6).
  • domain assumption The five control parameters and fixed geometry/transport profiles uniquely determine the steady-state fields in the sampled region.
    Inverse treats parameter recovery as well-posed; Sec. 3.2 notes ill-posedness and local insensitivity, so this is load-bearing.
  • domain assumption The 80/20 run-level split prevents leakage and held-out metrics generalize to new SOLPS-ITER runs.
    Relies on whole-run split only; with 608 training runs and 4.3M parameters, generalization is not guaranteed but is argued via held-out results.
  • domain assumption The committee ensemble variance is a valid proxy for predictive uncertainty.
    QBC method assumes disagreement tracks error; validated only indirectly through the inner-target flag pattern.
  • domain assumption The k-d tree warm-start completion rate can be compared with the cold-started ensemble of [22].
    The 29% cold-start failure rate is from a different study/setup; this comparison supports the claimed improvement.
  • domain assumption Uncertainty quality score scaling S computed on the validation fold transfers to the test set.
    The scaling uses validation-fold error and spread; test behavior assumes exchangeability.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas." pith.science (2026). https://pith.science/paper/IY5X5SH5

@misc{pith2026260721407,
  author       = {Pith},
  title        = {Pith review of: Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IY5X5SH5}},
  note         = {Machine review of arXiv:2607.21407}
}
abstract

The boundary and divertor plasma govern how a tokamak exhausts power and particles, setting heat fluxes, target conditions, and the onset of detachment. Predicting these quantities is essential for operating current and future devices, but edge simulations that resolve them are too slow for parameter scans, optimization, or real-time control. Machine-learning surrogates offer a fast alternative, yet most are forward-only: they cannot recover input parameters from observations or assess the reliability of their predictions. We introduce a cycle-consistent neural surrogate for edge plasmas, combining a conditional U-Net forward model with an optimization-based inverse method built on the frozen forward network. The forward model maps five control parameters to two-dimensional plasma-state fields on the SOLPS-ITER mesh; the inverse method enforces consistency between forward and inverse predictions, a self-supervised quality check needing no ground-truth labels at inference. An ensemble of multilayer perceptrons also predicts electron temperature and density profiles at the outboard midplane and divertor targets, with uncertainty estimates that flag where more simulations are needed. The forward model achieves normalized root-mean-square errors below 2.6% and Pearson correlations above 0.95 for all fields. Cycle-consistency regularization raises the average cyclical $R^2$ from 0.59 to 0.99 without degrading forward accuracy and enables recovery of the core fueling rate; all five control parameters are recovered with Pearson $r\ge0.97$. A $k$-d tree warm start yields a database completion rate above 95%, versus roughly 30% outright failures when cold-started. With about $4\times10^6$ parameters, the model produces full 2D predictions in milliseconds, five to six orders of magnitude faster than SOLPS-ITER, enabling real-time control, parameter scans, uncertainty analysis, and digital twins.

Figures

Figures reproduced from arXiv: 2607.21407 by the authors.

Figure 1
Figure 1. Surrogate model architecture: the three-stage inference cycle. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Training-time cycle-consistency regularizer. The forward model [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Comparison of SOLPS-ITER ground truth (left columns), U-Net predictions (middle columns), and absolute errors (right columns) [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Cycle-consistency ablation. (A) Average cyclical [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Inverse parameter recovery on held-out test cases. Each panel shows the true versus recovered value for one control parameter, [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Evaluation of the one-dimensional profile ensemble. Columns correspond to the three profile locations (upstream/outboard midplane, [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 5 canonical work pages

  1. [22]

    Dasbach, S

    S. Dasbach, S. Wiesen, Towards fast surrogate models for interpolation of tokamak edge plasmas, Nuclear Materials and Energy 34 (101396) (2023) 1–6.doi: http://dx.doi.org/10.1016/j.nme.2023.101396

    arXiv 2023

  2. [1]

    Wigley, P

    P. Wigley, P. Everitt, A. van den Hengel, J. Bas- tian, M. Sooriyabandara, G. McDonald, K. Hard- man, C. Quinlivan, P. Manju, C. Kuhn, I. Pe- terson, A. Luiten, J. Hope, N. Robins, M. Hush, Fast machine-learning online optimization of ultra- cold-atom experiments, Scientific Reports 6 (25890) (2016)

    2016

  3. [2]

    Scheinker, S

    A. Scheinker, S. Gessner, Adaptive method for elec- tron bunch profile prediction, Physical Review Accel- erators and Beams 18 (102801) (2015)

    2015

  4. [3]

    Noack, K

    M. Noack, K. Yager, M. Fukuto, G. Doerk, L. Ruipeng, J. Sethian, A kriging-based approach to autonomous experimentation with applications to x- ray scattering, Scientific Reports 9 (11809) (2019)

    2019

  5. [4]

    A. Diaw, M. McKerns, I. Sagert, L. G. Stanton, M. S. Murillo, Efficient learning of accurate surro- gates for simulations of complex systems, Nature Ma- chine Intelligence 6 (2024) 568–577.doi:10.1038/ s42256-024-00839-1

    2024

  6. [5]

    Schneider, X

    R. Schneider, X. Bonnin, K. Borrass, D. P. Coster, H. Kastelewicz, D. Reiter, V. A. Rozhansky, B. J. Braams, Plasma edge physics with B2-Eirene, Con- tributions to Plasma Physics 46 (1-2) (2006) 3–191. doi:10.1002/ctpp.200610001

    work page doi:10.1002/ctpp.200610001 2006

  7. [6]

    R. A. Pitts, X. Bonnin, F. Escourbiac, T. Hirai, V. Komarov, A. S. Kukushkin, A. Loarte, A. Mar- tin, M. Merola, R. Mitteau, Divertor and scrape- off layer plasma physics for reactor-grade fusion en- ergy devices, Nuclear Materials and Energy 20 (2019) 100696.doi:10.1016/j.nme.2019.100696

    arXiv 2019

  8. [7]

    Reiter, M

    D. Reiter, M. Baelmans, P. Börner, EIRENE – a Monte Carlo linear transport solver, Tech. rep., Forschungszentrum Jülich,http://www.eirene.de (2005)

    2005

Show all 45 references
  1. [8]

    Kates-Harbeck, A

    J. Kates-Harbeck, A. Svyatkovskiy, W. Tang, Ma- chine learning for disruption warnings on Alcator C- Mod, DIII-D, and EAST, Nature 568 (7753) (2019) 526–531.doi:10.1038/s41586-019-1116-4

    2019 doi

  2. [9]

    D. P. Kingma, M. Welling, An introduction to vari- ational autoencoders, Foundations and Trends®in Machine Learning 12 (4) (2019) 307–392.doi:10. 1561/2200000056. URLhttp://dx.doi.org/10.1561/2200000056

    2019 doi

  3. [10]

    Ronneberger, P

    O. Ronneberger, P. Fischer, T. Brox, U-net: Convo- lutional networks for biomedical image segmentation (2015).arXiv:1505.04597. URLhttps://arxiv.org/abs/1505.04597

    2015 arXiv

  4. [11]

    NN forward

    J. Degrave, F. Felici, J. Buchli, M. Neunert, B. Tracey, F. Carpanese, T. Ewalds, R. Hafner, 15 Figure C.9: Predicted electron profiles as a function of normalized poloidal fluxψN for a representative held-out test case.Solid lines: SOLPS-ITER ground truth.Dashed lines(“NN for...

    2022

  5. [12]

    J. Seo, S. Kim, A. Jalalvand, R. Conlin, A. Rothstein, J. Abbate, K. Erickson, J. Wai, R. Shousha, E. Kole- men, Avoiding fusion plasma tearing instability with deep reinforcement learning, Nature 626 (2024) 746– 751.doi:10.1038/s41586-024-07024-9

    2024 doi

  6. [13]

    W. Tang, E. Feibush, G. Dong, N. Borthwick, AI- machine learning-enabled tokamak digital twin, arXiv preprint (2024).arXiv:2409.03112

    2024 arXiv

  7. [14]

    Willcox, B

    K. Willcox, B. Segundo, The role of computa- tional science in digital twins, Nature Computa- tional Science 4 (2024) 147–149.doi:10.1038/ s43588-024-00609-4

    2024

  8. [15]

    A. Diaw, K. Barros, J. Haack, C. Junghans, B. Keenan, Y. W. Li, D. Livescu, N. Lubbers, M. McKerns, R. S. Pavel, D. Rosenberger, I. Sagert, T. C. Germann, Multiscale simulation of plasma flows using active learning, Phys. Rev. E 102 (2) (2020) 023310.doi:10.1103/PhysRevE.102.023310

    2020 doi

  9. [16]

    A. Diaw, M. McKerns, I. Sagert, L. G. Stanton, M. S. Murillo, Efficient Learning of Accurate Sur- rogates for Simulations of Complex Systems, arXiv 16 e-prints (2022) arXiv:2207.12855arXiv:2207.12855, doi:10.48550/arXiv.2207.12855

    2022 doi arXiv

  10. [17]

    Wiesen, D

    S. Wiesen, D. Reiter, V. Kotov, M. Baelmans, W. Dekeyser, A. S. Kukushkin, S. W. Lisgo, R. A. Pitts, V. Rozhansky, G. Saibene, I. Veselova, S. Voskoboynikov, The new solps-iter code pack- age, Journal of Nuclear Materials 463 (2015) 480– 484.doi:http://dx.doi.org/10.1016/j.jnu...

    2015 doi

  11. [18]

    Reiter, M

    D. Reiter, M. Baelmans, P. Borner, The eirene and b2-eirene codes, Fusion Science and Technology 47 (2) (2005) 172–186.doi:http://dx.doi.org/10. 13182/FST47-172

    2005

  12. [19]

    J. D. Lore, S. De Pascuale, P. Laiu, B. Russo, J. S. Park, J. M. Park, S. L. Brunton, J. N. Kutz, A. A. Kaptanoglu, Time-dependent solps-iter sim- ulations of the the tokamak plasma boundary for model predictive control using sindy, Nuclear Fusion 63 (046015) (2023) 1–12.doi:h...

    2023 doi

  13. [20]

    J. S. Park, J. D. Lore, M. Reinke, A. Q. Kuang, S. De Pascuale, A. Creely, Full time-dependent solps-iter simulation of the sparc tokamak: actuator design for particle and divertor condition control, Nuclear Fu- sion 64 (4) (2024) 1–14.doi:http://dx.doi.org/ 10.1088/1741-4326/ad4f9d

    2024 doi

  14. [23]

    Dasbach, A

    S.Wiesen, S. Dasbach, A. Kit, A. E. Jaervinen, A. Gillgren, A. Ho, A. Panera, D. Reiser, M. Bren- zke, Y. Poels, E. Westerhof, V. Menkovski, G. F. Derks, P. Strand, Data-driven models in fusion ex- haust: Ai methods and perspectives, Nuclear Fusion 64 (086046) (2024) 1–10.doi:...

    2024 doi

  15. [24]

    B. Zhu, M. Zhao, H. Bhatia, X. Xu, P. Bremer, W. Meyer, N. Li, T. Rognlien, Data-driven model for divertor plasma detachment prediction, Journal of Plasma Physics 88 (895880504) (2022) 1–23.doi: http://dx.doi.org/10.1017/S002237782200085X

    2022 doi

  16. [25]

    B. Zhu, M. Zhau, X. Xu, A. Gupta, K. B. Kwon, X. Ma, D. Eldon, Latent space mapping: Revolu- tionizing predictive models for divertor plasma de- tachment control, Physics of Plasmas 32 (062508) (2025) 1–19.doi:http://dx.doi.org/10.1063/5. 0267930

    2025 doi

  17. [26]

    Bentley, Multidimensional binary search trees used for associative searching, Communications of the ACM 18 (9) (1975) 509–517.doi:http://dx.doi

    J. Bentley, Multidimensional binary search trees used for associative searching, Communications of the ACM 18 (9) (1975) 509–517.doi:http://dx.doi. org/10.1145/361002.361007

    1975

  18. [27]

    Krizhevsky, I

    A. Krizhevsky, I. Sutskever, G. E. Hinton, Ima- genet classification with deep convolutional neural networks, Advances in neural information processing systems 25 (2012)

    2012

  19. [28]

    Howard, W

    Y.LeCun, B.Boser, J.S.Denker, D.Henderson, R.E. Howard, W. Hubbard, L. D. Jackel, Backpropagation applied to handwritten zip code recognition, Neural computation 1 (4) (1989) 541–551

    1989

  20. [29]

    Perez, F

    E. Perez, F. Strub, H. de Vries, V. Dumoulin, A. Courville, FiLM: Visual reasoning with a general conditioning layer, in: Proceedings of the AAAI Con- ference on Artificial Intelligence, Vol. 32, 2018, pp. 3942–3948.doi:10.1609/aaai.v32i1.11671. URLhttps://arxiv.org/abs/1709.07871

    2018 arXiv

  21. [30]

    D. P. Kingma, J. Ba, Adam: A method for stochastic optimization (2017).arXiv:1412.6980. URLhttps://arxiv.org/abs/1412.6980

    2017 arXiv

  22. [31]

    Paszke, S

    A. Paszke, S. Gross, F. Massa, A. Lerer, J. Brad- bury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. Köpf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, S. Chintala, Pytorch: An imperative style, high-perfo...

    2019 arXiv

  23. [32]

    Transfer Learning as a Tool for Reducing Simula- tion Bias: Application to Inertial Confinement Fu- sion

    B. Kustowski, J. A. Gaffney, B. K. Spears, G. J. An- derson, J. J. Thiagarajan, R. Anirudh, Erratum to “Transfer Learning as a Tool for Reducing Simula- tion Bias: Application to Inertial Confinement Fu- sion” [Jan 20 46-53], IEEE Transactions on Plasma Science 48 (9) (2020) 3...

    2020

  24. [33]

    Anirudh, J

    R. Anirudh, J. J. Thiagarajan, P.-T. Bremer, B. K. Spears, Improved surrogates in inertial confinement fusion withmanifold and cycleconsistencies, Proceed- ings of the National Academy of Sciences 117 (18) (2020) 9741–9746.doi:10.1073/pnas.1916634117

    2020 doi

  25. [34]

    J. St. John, C. Herwig, D. Kafkes, J. Mitrevski, W. A. Pellico, G. N. Perdue, A. Quintero-Parra, B. A. Schupbach, K. Seiya, N. Tran, M. Schram, 17 J. M. Duarte, Y. Huang, R. Keller, Real-time ar- tificial intelligence for accelerator control: A study at the Fermilab Booster, P...

    2021

  26. [35]

    H. S. Seung, M. Opper, H. Sompolinsky, Query by committee, in: ProceedingsoftheFifthAnnualWork- shop on Computational Learning Theory, COLT ’92, Association for Computing Machinery, New York, NY, USA, 1992, p. 287–294.doi:10.1145/130385. 130417. URLhttps://doi.org/10.1145/1303...

    1992

  27. [36]

    R. S. Wilcox, M. W. Shafer, J. D. Lore, J. M. Canik, S. R. Haskey, C. J. Lasnier, A. L. Moser, T. H. Os- borne, H. Q. Wang, Challenges and approaches to interpretive modeling of boundary plasma and neu- tral transport in a closed, pumped divertor, Nuclear Fusion 66 (2026) 0160...

    2026 doi

  28. [37]

    Roussel, A

    R. Roussel, A. Edelen, C. Mayes, D. Ratner, J. P. Gonzalez-Aguilera, S. Kim, E. Wisniewski, J. Power, Phase space reconstruction from accelerator beam measurements using neural networks and differen- tiablesimulations, PhysicalReviewLetters130(2023) 145001.doi:10.1103/PhysRevL...

    2023 doi

  29. [38]

    E. L. Allgower, K. Georg, Introduction to Numeri- cal Continuation Methods, Vol. 45 of Classics in Ap- plied Mathematics, SIAM, 2003.doi:10.1137/1. 9780898719154

    2003 doi

  30. [39]

    M. D. McKay, R. J. Beckman, W. J. Conover, A com- parison of three methods for selecting values of in- put variables in the analysis of output from a com- puter code, Technometrics 21 (2) (1979) 239–245. doi:10.1080/00401706.1979.10489755

    1979

  31. [40]

    J. H. Friedman, J. L. Bentley, R. A. Finkel, An algorithm for finding best matches in logarithmic expected time, ACM Transactions on Mathemati- cal Software 3 (3) (1977) 209–226.doi:10.1145/ 355744.355745

    1977

  32. [41]

    M. P. Allen, D. J. Tildesley, Computer Simulation of Liquids, 2nd Edition, Oxford University Press, 2017. doi:10.1093/oso/9780198803195.001.0001

    2017

  33. [42]

    Frenkel, B

    D. Frenkel, B. Smit, Understanding Molecular Sim- ulation: From Algorithms to Applications, 2nd Edi- tion, Academic Press, 2002

    2002

  34. [43]

    A. Diaw, M. S. Murillo, Generalized hydrodynamics model for strongly coupled plasmas, Phys. Rev. E 92 (1) (2015) 013107.doi:10.1103/PhysRevE.92. 013107

    2015 doi

  35. [44]

    Ticknor, J

    C. Ticknor, J. D. Kress, L. A. Collins, J. Clérouin, P. Arnault, A. Decoster, Transport properties of an asymmetric mixture in the dense plasma regime, Phys. Rev. E 93 (6) (2016) 063208.doi:10.1103/ PhysRevE.93.063208

    2016

  36. [45]

    Flyvbjerg, H

    H. Flyvbjerg, H. G. Petersen, Error estimates on av- erages of correlated data, The Journal of Chemical Physics 91 (1989) 461–466.doi:10.1063/1.457480

    1989 doi

  37. [46]

    A. D. Sokal, Monte Carlo methods in statistical me- chanics: Foundations and new algorithms, Lecture Notes in Physics 490 (1997) 131–192. 18

    1997

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.