Pith. sign in

REVIEW 3 major objections 5 minor 76 references

Automated simulation-based design via multi-fidelity active learning and optimisation for laser direct drive implosions

T0 review · 3 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read A machine-learning design loop that leans on cheap 1D simulations finds a laser-fusion target which, hydrodynamically scaled from 25 kJ to 2 MJ, burns with 217-fold yield amplification instead of collapsing under seeded instabilities.

desk verdict Worth reading for the 25 kJ multi-fidelity design loop; don't quote the 2 MJ high-gain numbers without the hydro-scaling caveat. read the letter →

arxiv 2508.20878 v1 pith:NWJXHMPL submitted 2025-08-28 physics.plasm-ph

A. J. Crilly , P. W. Moloney , D. Shi , E. A. Ferdinandi This is my paper · ORCID
classification physics.plasm-ph
keywords inertialconfinementfusionlaserdirectdrivemulti-fidelitysurrogatemodelsactivelearningBayesianoptimisationhydrodynamicinstabilitiesneuralnetworkensemblescaling
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's aim is to make laser-direct-drive fusion design resilient to hydrodynamic instabilities without paying the full cost of multidimensional simulations. It builds a neural-network surrogate that first learns the good-design region from roughly 12,500 cheap 1D simulations, then transfers that knowledge to a small set of about 128 expensive 2D simulations that seed beam-mode and ablator-surface perturbations. Bayesian optimisation on the 2D-trained surrogate identifies an eight-parameter design at 25 kJ that is only mildly degraded by instabilities, whereas the design optimised in 1D loses its shell. Hydrodynamically scaled 2D simulations at 2 MJ, with alpha heating on, give about 2.2e19 neutrons, 217x yield amplification, and roughly 15% burn fraction for the 2D-optimised design, versus 1.0e18 neutrons, 17x amplification, and 0.9% burn for the 1D-optimised design. The core claim is that instabilities can be designed against at small scale using an information-transfer surrogate, and that the resulting target ignites and propagates burn when scaled up.

What carries the argument

The load-bearing mechanism is a multi-fidelity neural-network ensemble trained with transfer learning: a 25-member multilayer-perceptron ensemble learns the 1D design landscape, then all but the last two layers are frozen and the remaining layers are retrained on 2D data, so knowledge of where good designs live transfers across fidelities. Around it sits a pipeline of quasi-random initial sampling, probabilistic-threshold active learning to concentrate 2D runs in promising regions, and Bayesian optimisation with a log Expected Improvement acquisition function using the multi-fidelity surrogate at fixed fidelity. The objective function couples the chi_no_alpha ignition criterion to areal dens

What would settle it

Run the O2D design at 2 MJ with a properly hydro-equivalent re-tuned pulse and target, for example exchanging ablator mass for DT ice as hydro-equivalent ignition theory prescribes, and compare the burn-on yield against the reported 217x amplification; also fire the O2D target on a 25 kJ laser with seeded surface perturbations and compare yield and areal density with the 2D predictions. Agreement would confirm the scaled result, while a collapse toward the 1D-optimised design's 17x amplification would falsify it.

Watch

Extended reading notes

Core claim

The central discovery is that the space of designs resilient to hydrodynamic instabilities can be learned almost entirely from 1D simulations, with a modest number of 2D simulations used as transfer data. The authors define a single scalar objective that blends the no-alpha ignition metric chi_no_alpha with post-ignition burn propagation estimated through areal density, scaled by the hydrodynamic scale factor S from 25 kJ to 2 MJ. An ensemble of 25 multilayer-perceptron surrogates trained on the 1D dataset, with all but the last two layers frozen and retrained on the 2D dataset, provides calibrated uncertainties for active learning and Bayesian optimisation. The resulting 2D-optimised design

Load-bearing premise

The 2 MJ high-gain numbers rest on the assumption that a hydrodynamically equivalent implosion can be produced up to stagnation by simple scaling, without re-tuning the laser pulse and target; if that fails, the scaled yields are extrapolations rather than confirmations.

Editorial extensions

If this is right

  • A 2D-optimised design at 25 kJ is substantially more stable than the 1D-optimised one: peak inflight aspect ratio 30 versus 37, higher shell adiabat, and an intact shell at bang time.
  • Hydro-scaling to 2 MJ in 2D with burn-on gives the 2D-optimised design about 2.24e19 neutrons, yield amplification 217, and about 15% burn fraction, versus 1.02e18 neutrons, 17x, and 0.9% for the 1D-optimised design.
  • The framework locates resilient designs with only about 128 2D simulations because 1D transfer narrows the search space dramatically.
  • The optimiser automatically found shock-timing and stability tradeoffs, such as a higher picket power and thicker ice, without being explicitly asked to tune those quantities.
  • The chosen objective rewards designs that would ignite and propagate burn at a larger scale rather than merely maximising 25 kJ yield.

Reading between the lines

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

  • If the hydro-equivalence assumption fails at 2 MJ, the reported O2D yield is an extrapolation; the design should be re-tuned with a full hydro-equivalent pulse and target, then re-simulated, before high gain is treated as confirmed.
  • The same transfer-learning loop could be pointed at 3D simulations, experimental data, or additional perturbation sources such as ice roughness and stalk shadows; each would likely shift the resilient design region the surrogate finds.
  • A direct 25 kJ laser experiment firing the O2D target with seeded surface perturbations would test whether the predicted resilience appears in measured yield and areal density; agreement would strengthen the scaled 2 MJ claim.
  • Amplifying the seeded ablator perturbation level should push the optimised design toward smaller capsules, thicker ice, and higher picket power, as the authors note; this trend is a testable prediction of the surrogate's learned physics.
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 / 5 minor

Summary. The paper presents an automated, multi-fidelity simulation-based design framework for laser direct drive inertial confinement fusion. The authors use ~12.5k 1D Chimera simulations to train an ensemble of MLP surrogates, transfer learned features to a much smaller 2D dataset, and then run active learning (probabilistic threshold sampling) and Bayesian optimisation (log expected improvement) over an 8-parameter, OMEGA-relevant design space at 25 kJ. The objective Y (Eq. 3a) combines a hydro-scaled, no-alpha ignition metric with an areal-density term for burn propagation. The 2D-optimised design O2D is reported to be more hydrodynamically stable than the 1D-optimised design O1D at 25 kJ. The paper then hydro-scales the 25 kJ 2D restart fields to 2 MJ and runs 2D burn-off/burn-on simulations, reporting for O2D a yield of 2.24e19 neutrons, yield amplification ~217, and ~15% burn fraction, versus 1.02e18 neutrons and amplification ~17 for O1D.

Significance. If the 25 kJ results are taken at face value, the paper gives a useful demonstration that a multi-fidelity surrogate, active learning, and Bayesian optimisation loop can identify a design that is more robust in 2D than a purely 1D-optimised design while using only a modest number of expensive 2D runs. The open-source orchestration package, the calibrated NN ensembles, and the explicit statement of the hydro-scaling caveat are strengths. However, the headline 2 MJ high-gain claim is not an independent confirmation: it rests on the assumption that a hydro-equivalent implosion can be achieved up to a restart time near stagnation, which the authors explicitly do not test. This makes the central claim conditional, and the conclusion as currently worded overstates the evidence. The paper is a reasonable candidate for publication after the high-gain claim is re-framed and the uncertainty of the 2 MJ results is addressed.

major comments (3)
  1. [Section IV B / Table II] The 2 MJ high-gain result (O2D: 2.24e19 neutrons, yield amplification 217, ~15% burn fraction) is not a direct confirmation of high gain. The authors state: "we will not perform this re-optimisation but instead apply hydro-scaling to the 2D simulations at a restart time approaching stagnation, making the assumption that a hydro-equivalent implosion can be achieved up to this time." The table footnote repeats that full hydro-equivalent re-tuning was not performed. Nora et al. (Ref. 19), cited by the authors, show that restoring hydro-equivalence requires additional target and pulse changes, mainly exchanging ablator mass for DT ice. Since the conclusion in Section VI says the framework "correctly identified a high-gain scaled-up design," that claim is stronger than the simulation evidence. The stress-test concern therefore lands. I recommend either running a full 2 MJ simulation from the
  2. [Section IV B / Table II] The 2 MJ burn-on and burn-off values are single 2D simulations for each design, with no error bars or seed-to-seed statistics. This matters because the 25 kJ 2D value of chi_S,no_alpha for O2D is 1.72, close to the nominal ignition threshold, and the 2D simulations intentionally use randomized ablator density perturbations to represent shot-to-shot variability. A different random seed at the 2 MJ scale could plausibly change whether burn propagates and could move the yield by orders of magnitude. The robustness claim therefore needs at least a small ensemble (e.g., 3-5 perturbation realizations) at 2 MJ, or a quantitative propagation of the 25 kJ seed distribution through the scaling procedure. Without this, the 2 MJ performance numbers should be treated as a single realization, not a robust prediction.
  3. [Section II C 2 / Fig. 1] Quantitative validation of the 2D surrogate on held-out 2D data is not reported. Fig. 1(b)(iii) shows a before/after comparison for the transfer-learned 2D model but gives no R^2, RMSE, or coverage numbers for a 2D test set. Fig. 4 compares 1D and 2D surrogates on inputs from the 1D database, not on held-out 2D simulations. Since the active learning and Bayesian optimisation decisions are driven by the 2D ensemble's mean and calibrated uncertainty, the paper should include a table or plot with held-out 2D prediction error and calibration statistics, ideally in the region around the reported optima. This would let the reader assess how much the 25 kJ and consequently 2 MJ conclusions depend on surrogate accuracy.
minor comments (5)
  1. [Section II A, Eq. (4c)] The notation d^2 Y_DT / (dt dV) is non-standard and dimensionally confusing. Please write the double integral explicitly (over time and volume) or use a clearer mixed-derivative notation.
  2. [Section II B] "2.5um" should be "2.5 μm".
  3. [Section IV B] The sentence "First, both designs have sufficient ignition margin to ignite in 2D at the 2 MJ energy scale" is difficult to reconcile with the O1D result, which is later described as failing to propagate burn into the dense fuel. Recommend distinguishing hot-spot ignition from propagating burn, e.g., "sufficient margin for hot-spot ignition but insufficient confinement for propagating burn."
  4. [Fig. 1(b)(iii)] The caption says "2D dataset" but does not state whether the points are a held-out 2D test set or the training set. Please clarify and include quantitative performance metrics on the figure.
  5. [Abstract / Conclusions] The phrase "confirm the achievement of high gain" and similar wording in the conclusion is stronger than the conditional statement in Section IV B. If the major comment about hydro-equivalence is addressed by re-framing, the abstract and conclusions should be aligned with the conditional framing.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the 2 MJ high-gain claim is an explicitly qualified hydro-scaling extrapolation, not an equation-level self-derivation.

full rationale

The paper's derivation chain is not circular. The 25 kJ optimization is driven by the physics-informed objective Eq. (3a), which combines the hydro-scaled ignition metric χ_{S,noα} (Eq. 3b) with an areal-density burn-propagation term. The 2 MJ results in Table II are obtained from full 2D Chimera radiation-hydrodynamics burn-on simulations after applying the scaling R→SR, t→St, E→S^3E (Sec. IV B); they are not evaluations of Eq. (3a), nor re-statements of the surrogate predictions. This is why O1D and O2D, which have comparable 1D objective values, differ dramatically in the 2D burn-on yield (1.02e18 vs 2.24e19 neutrons). The load-bearing limitation is explicitly acknowledged: 'we will not perform this re-optimisation but instead apply hydro-scaling to the 2D simulations at a restart time approaching stagnation, making the assumption that a hydro-equivalent implosion can be achieved up to this time.' This is a physical assumption about the validity of hydro-equivalence without re-tuning, which is a correctness/robustness concern, not an instance of a predicted quantity being equal to an input by construction. The self-citations (Chimera, SpK, Tong et al. burn package) are normal code/method references and are not used to forbid alternatives or to import an unverified uniqueness claim. No fitted parameter is renamed as a prediction, and no equation reduces to another by definition. The paper therefore merits a circularity score of 0, with the caveat that the 2 MJ high-gain claim is only as strong as the stated hydro-equivalence assumption.

Assumptions & free parameters 6 free parameters · 8 assumptions · 0 invented entities

No physical constants are fitted in this paper, but the reported optimum and the 2 MJ confirmation depend on hand-chosen design bounds, perturbation amplitude, active-learning thresholds, objective normalization, and especially the hydro-equivalence assumption. The chi and rhoR exponents are imported from cited literature and counted as domain assumptions rather than free fits. No new particles, forces, or conserved quantities are introduced.

free parameters (6)
  • Ensemble uncertainty calibration factor c_v = ~3.5
    Appendix A; fit to make ensemble coverage match empirical frequency. Its value affects acquisition functions and where Bayesian optimization looks, though not the rad-hydro outputs themselves.
  • Active-learning thresholds and P_min = Y_threshold=1 (1D), 2 (2D); P_min=0.25 to 0.5
    Algorithm 1; chosen by hand to balance exploration and exploitation. These choices determine the distribution of 2D samples and can change which optimum is found.
  • Ablator density perturbation amplitude = ~1% RMS areal density variation in initial conditions
    Section II B; chosen by hand as the instability seed. It defines what robustness to hydrodynamic instabilities means in this study, and the authors note that larger seeds would push optima toward thicker ice and higher picket power.
  • MLP hyperparameters = hidden layers 32,48,48,32; learning rate 0.01 halving every 20 epochs; weight decay 0.05; 80 to 100 epochs; ensemble siz
    Section II C 2; selected by manual and automated tuning. They affect surrogate accuracy and therefore optimization trajectories.
  • Objective normalization rho_R_hat = 0.1 g/cm2
    Eq. 3d; chosen to scale the areal-density bonus term in the scalar objective. It directly changes the ranking of candidate designs.
  • Design-space bounds and pulse parameterization = Table I; pulse durations capped at 1 ns; sigma_t=25 ps; beta=9/5
    These choices define the optimization domain. The reported optima are only optima within this box; changing bounds or the functional pulse form would change the results.
assumptions (8)
  • domain assumption Chimera/SOLAS radiation-hydrodynamics simulations faithfully represent 1D and 2D LDD implosions, including alpha heating and burn at the 2 MJ scale.
    Section II B; all yields, areal densities, temperatures, and burn fractions come from these codes, with no independent experimental verification.
  • domain assumption The seeded perturbations, OMEGA beam mode and randomized density perturbations on the outer 2.5 um of the CH ablator, are a sufficient proxy for hydrodynamic instability sources.
    Section II B; Section V lists ice roughness, stalk, and other perturbation sources as future work, so robustness claims are limited to the included seeds.
  • ad hoc to paper Hydrodynamic scaling without re-optimization gives a representative 2 MJ implosion.
    Section IV B; the authors explicitly assume a hydro-equivalent implosion can be achieved up to stagnation and do not perform the Nora et al. re-tuning. This is the main load-bearing caveat for the 2 MJ confirmation.
  • domain assumption The ignition metric chi_no_alpha and the Fraley burn fraction formula are valid predictors for sub-scale designs.
    Section II A; the exponents 0.61, 0.34, -1.64 and the 6.3 g/cm2 denominator are imported from Betti, Christopherson, and Fraley prior work.
  • ad hoc to paper The scalar objective Y (Eq. 3a) is an adequate proxy for scaled ignition and high gain.
    Section II A; this hybrid objective was devised for this study and is only checked by the two 2 MJ simulation pairs at the end.
  • domain assumption The 25-member neural-network ensemble, after transfer learning and uncertainty calibration, is accurate enough for active learning and Bayesian optimization to locate near-global optima in the 8D design box.
    Section II C and D; no convergence guarantees are given, only validation R2 and calibration curves on withheld data.
  • domain assumption The automatic restart from 2D spherical to 2D cylindrical geometry does not introduce numerical artifacts that dominate the instability seed.
    Section II B; the restart is designed to avoid polar-axis time-step restrictions, but no grid-convergence study is reported.
  • standard math Bayesian optimization and Gaussian process/neural-network probability models as implemented in BOTorch are correct and converged.
    Section II E; these are standard methods, but the acquisition-function optimization is numerical and no convergence certificates are provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Automated simulation-based design via multi-fidelity active learning and optimisation for laser direct drive implosions." pith.science (2026). https://pith.science/paper/NWJXHMPL

@misc{pith2026250820878,
  author       = {Pith},
  title        = {Pith review of: Automated simulation-based design via multi-fidelity active learning and optimisation for laser direct drive implosions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NWJXHMPL}},
  note         = {Machine review of arXiv:2508.20878}
}
read the original abstract

The design of inertial fusion experiments is a complex task as driver energy must be delivered in a precise manner to a structured target to achieve a fast, but hydrodynamically stable, implosion. Radiation-hydrodynamics simulation codes are an essential tool in this design process. However, multi-dimensional simulations that capture hydrodynamic instabilities are more computationally expensive than optimistic, 1D, spherically symmetric simulations which are often the primary design tool. In this work, we develop a machine learning framework that aims to effectively use information from a large number of 1D simulations to inform design in the presence of hydrodynamic instabilities. We use an ensemble of neural network surrogate models trained on both 1D and 2D data to capture the space of good designs, i.e. those that are robust to hydrodynamic instabilities. We use this surrogate to perform Bayesian optimisation to find optimal designs for a 25 kJ laser driver. We perform hydrodynamic scaling on these designs to confirm the achievement of high gain for a 2 MJ laser driver, using 2D simulations including alpha heating effects.

Figures

Figures reproduced from arXiv: 2508.20878 by the authors.

Figure 1
Figure 1. FIG. 1. (a) Shows the final architecture chosen for the multi [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. A schematic showing the ML and simulation workflow [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. (Top) History of objective values obtained [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (7 more)
Figure 5
Figure 5. Figure 5: FIG. 5. (Top) History of objective values obtained [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 7
Figure 7. Figure 7: FIG. 7. (Top) Time series of the inflight aspect ratio (IFAR) [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Bang time conditions (mass density [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Neural network ensemble surrogate model predic [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10. Time series of key burn parameters: burn rate (DT [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. Bang time conditions (mass density [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12. Plot of expected frequency vs observed frequency, [PITH_FULL_IMAGE:figures/full_fig_p013_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

76 extracted references · 70 canonical work pages

  1. [1]

    The Physics of Inertial Fusion: BeamPlasma Interaction, Hydrodynamics, Hot Dense Matter

    Stefano Atzeni and J \" u rgen Meyer-ter Vehn. The Physics of Inertial Fusion: BeamPlasma Interaction, Hydrodynamics, Hot Dense Matter . Oxford University Press, Oxford, 2004

    work page 2004

  2. [2]

    Development of the indirect-drive approach to inertial confinement fusion and the target physics basis for ignition and gain

    John Lindl. Development of the indirect-drive approach to inertial confinement fusion and the target physics basis for ignition and gain. Physics of plasmas , 2(11):3933--4024, 1995

    work page 1995

  3. [3]

    Direct-drive inertial confinement fusion: A review

    RS Craxton, KS Anderson, TR Boehly, VN Goncharov, DR Harding, JP Knauer, RL McCrory, PW McKenty, DD Meyerhofer, JF Myatt, et al. Direct-drive inertial confinement fusion: A review. Physics of Plasmas , 22(11), 2015

    work page 2015

  4. [4]

    Lawson criterion for ignition exceeded in an inertial fusion experiment

    H Abu-Shawareb, R Acree, P Adams, J Adams, B Addis, R Aden, P Adrian, BB Afeyan, M Aggleton, L Aghaian, et al. Lawson criterion for ignition exceeded in an inertial fusion experiment. Physical Review Letters , 129(7):075001, 2022

    work page 2022

  5. [5]

    A. B. Zylstra, A. L. Kritcher, O. A. Hurricane, D. A. Callahan, J. E. Ralph, D. T. Casey, A. Pak, O. L. Landen, B. Bachmann, K. L. Baker, L. Berzak Hopkins, S. D. Bhandarkar, J. Biener, R. M. Bionta, N. W. Birge, T. Braun, T. M. Briggs, P. M. Celliers, H. Chen, C. Choate, D. S. Clark, L. Divol, T. D\"oppner, D. Fittinghoff, M. J. Edwards, M. Gatu Johnson,...

    work page 2022

  6. [6]

    A. L. Kritcher, A. B. Zylstra, D. A. Callahan, O. A. Hurricane, C. R. Weber, D. S. Clark, C. V. Young, J. E. Ralph, D. T. Casey, A. Pak, O. L. Landen, B. Bachmann, K. L. Baker, L. Berzak Hopkins, S. D. Bhandarkar, J. Biener, R. M. Bionta, N. W. Birge, T. Braun, T. M. Briggs, P. M. Celliers, H. Chen, C. Choate, L. Divol, T. D\"oppner, D. Fittinghoff, M. J....

    work page 2022

  7. [7]

    Inertial-confinement fusion with lasers

    R Betti and OA Hurricane. Inertial-confinement fusion with lasers. Nature Physics , 12(5):435--448, 2016

    work page 2016

  8. [8]

    O. A. Hurricane , D. A. Callahan , D. T. Casey , P. M. Celliers , C. Cerjan , E. L. Dewald , T. R. Dittrich , T. D \"o ppner , D. E. Hinkel , L. F. B. Hopkins , J. L. Kline , S. Le Pape , T. Ma , A. G. Macphee , J. L. Milovich , A. Pak , H.-S. Park , P. K. Patel , B. A. Remington , J. D. Salmonson , P. T. Springer , and R. Tommasini . Fuel gain exceeding ...

    work page 2014

Show all 76 references
  1. [9]

    Cross-code comparison of the impact of the fill tube on high yield implosions on the national ignition facility

    Brian M Haines, Daniel S Clark, Christopher R Weber, M John Edwards, Steven H Batha, and John L Kline. Cross-code comparison of the impact of the fill tube on high yield implosions on the national ignition facility. Physics of Plasmas , 27(8), 2020

    2020

  2. [10]

    D. S. Clark, C. R. Weber, J. L. Milovich, A. E. Pak, D. T. Casey, B. A. Hammel, D. D. Ho, O. S. Jones, J. M. Koning, A. L. Kritcher, M. M. Marinak, L. P. Masse, D. H. Munro, M. V. Patel, P. K. Patel, H. F. Robey, C. R. Schroeder, S. M. Sepke, and M. J. Edwards. Three-dimension...

    2019

  3. [11]

    Diagnostic signatures of performance degrading perturbations in inertial confinement fusion implosions

    K McGlinchey, BD Appelbe, AJ Crilly, JK Tong, CA Walsh, and JP Chittenden. Diagnostic signatures of performance degrading perturbations in inertial confinement fusion implosions. Physics of Plasmas , 25(12):122705, 2018

    2018

  4. [12]

    Practical bayesian optimization of machine learning algorithms

    Jasper Snoek, Hugo Larochelle, and Ryan P Adams. Practical bayesian optimization of machine learning algorithms. Advances in neural information processing systems , 25, 2012

    2012

  5. [13]

    Bayesian calibration of computer models

    Marc C Kennedy and Anthony O'Hagan. Bayesian calibration of computer models. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 63(3):425--464, 2001

    2001

  6. [14]

    Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling

    Paris Perdikaris, Maziar Raissi, Andreas Damianou, Neil D Lawrence, and George Em Karniadakis. Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences , 473(2198):2...

    2017

  7. [15]

    Alpha heating and burning plasmas in inertial confinement fusion

    R Betti, AR Christopherson, BK Spears, R Nora, A Bose, J Howard, KM Woo, MJ Edwards, and J Sanz. Alpha heating and burning plasmas in inertial confinement fusion. Physical review letters , 114(25):255003, 2015

    2015

  8. [16]

    Theory of ignition and burn propagation in inertial fusion implosions

    AR Christopherson, R Betti, S Miller, V Gopalaswamy, OM Mannion, and D Cao. Theory of ignition and burn propagation in inertial fusion implosions. Physics of Plasmas , 27(5), 2020

    2020

  9. [17]

    G. S. Fraley, E. J. Linnebur, R. J. Mason, and R. L. Morse. Thermonuclear burn characteristics of compressed deuterium‐tritium microspheres. The Physics of Fluids , 17(2):474--489, 1974

    1974

  10. [18]

    Initial performance results of the omega laser system

    TR Boehly, DL Brown, RS Craxton, RL Keck, JP Knauer, JH Kelly, TJ Kessler, SA Kumpan, SJ Loucks, SA Letzring, et al. Initial performance results of the omega laser system. Optics communications , 133(1-6):495--506, 1997

    1997

  11. [19]

    R. Nora, R. Betti, K. S. Anderson, A. Shvydky, A. Bose, K. M. Woo, A. R. Christopherson, J. A. Marozas, T. J. B. Collins, P. B. Radha, S. X. Hu, R. Epstein, F. J. Marshall, R. L. McCrory, T. C. Sangster, and D. D. Meyerhofer. Theory of hydro-equivalent ignition for inertial fu...

    2014

  12. [20]

    The national ignition facility

    George H Miller, Edward I Moses, and Craig R Wuest. The national ignition facility. Optical Engineering , 43(12):2841--2853, 2004

    2004

  13. [21]

    Automated and highly parallelized bayesian optimization scheme for direct drive fusion experiments on omega

    V Gopalaswamy, A Lees, R Ejaz, CA Thomas, TJB Collins, KS Anderson, W Ebmeyer, and R Betti. Automated and highly parallelized bayesian optimization scheme for direct drive fusion experiments on omega. Physical Review Research , 7(1):013009, 2025

    2025

  14. [22]

    The blind implosion-maker: Automated inertial confinement fusion experiment design

    PW Hatfield, SJ Rose, and RHH Scott. The blind implosion-maker: Automated inertial confinement fusion experiment design. Physics of Plasmas , 26(6):062706, 2019

    2019

  15. [23]

    J. P. Chittenden, B. D. Appelbe, F. Manke, K. McGlinchey, and N. P.L. Niasse. Signatures of asymmetry in neutron spectra and images predicted by three-dimensional radiation hydrodynamics simulations of indirect drive implosions . Physics of Plasmas , 23(5), 2016

    2016

  16. [24]

    C. A. Walsh, J. P. Chittenden, K. McGlinchey, N. P.L. Niasse, and B. D. Appelbe. Self-Generated Magnetic Fields in the Stagnation Phase of Indirect-Drive Implosions on the National Ignition Facility . Physical Review Letters , 118(15):1--5, 2017

    2017

  17. [25]

    J.K. Tong, K. McGlinchey, B.D. Appelbe, C.A. Walsh, A.J. Crilly, and J.P. Chittenden. Burn regimes in the hydrodynamic scaling of perturbed inertial confinement fusion hotspots. Nuclear Fusion , 59(8):086015, jun 2019

    2019

  18. [26]

    Neutron backscatter edges as a diagnostic of burn propagation

    AJ Crilly, BD Appelbe, OM Mannion, CJ Forrest, JP Knauer, DJ Schlossberg, EP Hartouni, AS Moore, and JP Chittenden. Neutron backscatter edges as a diagnostic of burn propagation. Physics of Plasmas , 29(6):062707, 2022

    2022

  19. [27]

    Spk: A fast atomic and microphysics code for the high-energy-density regime

    AJ Crilly, NPL Niasse, AR Fraser, DA Chapman, KM McLean, SJ Rose, and JP Chittenden. Spk: A fast atomic and microphysics code for the high-energy-density regime. arXiv preprint arXiv:2211.16464 , 2022

    2022 arXiv

  20. [28]

    Journal of Computational Physics , 257:594--626, 2014

    A stabilized runge–kutta–legendre method for explicit super-time-stepping of parabolic and mixed equations. Journal of Computational Physics , 257:594--626, 2014

    2014

  21. [29]

    Uniformity of energy deposition for laser driven fusion

    Stanley Skupsky and Kotik Lee. Uniformity of energy deposition for laser driven fusion. Journal of Applied Physics , 54(7):3662--3671, 1983

    1983

  22. [30]

    Using statistical modeling to predict and understand fusion experiments

    V Gopalaswamy, R Betti, JP Knauer, A Lees, D Patel, AR Christopherson, IV Igumenshchev, D Cao, KS Anderson, A Shvydky, et al. Using statistical modeling to predict and understand fusion experiments. Physics of Plasmas , 28(12), 2021

    2021

  23. [31]

    Irradiation uniformity for high-compression laser-fusion experiments

    S Skupsky and RS Craxton. Irradiation uniformity for high-compression laser-fusion experiments. Physics of Plasmas , 6(5):2157--2163, 1999

    1999

  24. [32]

    A model of laser imprinting

    VN Goncharov, S Skupsky, TR Boehly, JP Knauer, P McKenty, VA Smalyuk, RPJ Town, OV Gotchev, R Betti, and DD Meyerhofer. A model of laser imprinting. Physics of Plasmas , 7(5):2062--2068, 2000

  25. [33]

    Effects of laser bandwidth in direct-drive high-performance dt-layered implosions on the omega laser

    Dhrumir Patel, JP Knauer, D Cao, R Betti, Ryan Nora, A Shvydky, V Gopalaswamy, A Lees, Siddharth Sampat, WR Donaldson, et al. Effects of laser bandwidth in direct-drive high-performance dt-layered implosions on the omega laser. Physical Review Letters , 131(10):105101, 2023

    2023

  26. [34]

    Linear stability analysis of laser-driven spherical implosions

    JN Shiau, EB Goldman, and CI Weng. Linear stability analysis of laser-driven spherical implosions. Physical Review Letters , 32(7):352, 1974

    1974

  27. [35]

    A multifidelity bayesian optimization method for inertial confinement fusion design

    Jingyi Wang, N Chiang, Andrew Gillette, and J Luc Peterson. A multifidelity bayesian optimization method for inertial confinement fusion design. Physics of Plasmas , 31(3), 2024

    2024

  28. [36]

    The design and analysis of computer experiments , volume 1

    Thomas J Santner, Brian J Williams, William I Notz, and Brain J Williams. The design and analysis of computer experiments , volume 1. Springer, 2003

    2003

  29. [37]

    Gaussian processes for machine learning , volume 2

    Christopher KI Williams and Carl Edward Rasmussen. Gaussian processes for machine learning , volume 2. MIT press Cambridge, MA, 2006

    2006

  30. [38]

    The kernel cookbook: Advice on covariance functions

    David Duvenaud. The kernel cookbook: Advice on covariance functions. URL https://www. cs. toronto. edu/duvenaud/cookbook , 2014

    2014

  31. [39]

    Predicting fusion ignition at the national ignition facility with physics-informed deep learning

    Brian K Spears, Scott Brandon, Dan T Casey, John E Field, Jim A Gaffney, Kelli D Humbird, Andrea L Kritcher, Michael KG Kruse, Eugene Kur, Bogdan Kustowski, et al. Predicting fusion ignition at the national ignition facility with physics-informed deep learning. Science , 389(6...

    2025

  32. [40]

    Deep learning-based predictive models for laser direct drive at the omega laser facility

    Rahman Ejaz, Varchas Gopalaswamy, A Lees, C Kanan, D Cao, and R Betti. Deep learning-based predictive models for laser direct drive at the omega laser facility. Physics of Plasmas , 31(5), 2024

    2024

  33. [41]

    Data-driven prediction of scaling and ignition of inertial confinement fusion experiments

    Jim A Gaffney, Kelli Humbird, Andrea Kritcher, Michael Kruse, Eugene Kur, Bogdan Kustowski, Ryan Nora, and Brian Spears. Data-driven prediction of scaling and ignition of inertial confinement fusion experiments. Physics of Plasmas , 31(9), 2024

    2024

  34. [42]

    Simple and scalable predictive uncertainty estimation using deep ensembles

    Balaji Lakshminarayanan, Alexander Pritzel, and Charles Blundell. Simple and scalable predictive uncertainty estimation using deep ensembles. Advances in neural information processing systems , 30, 2017

    2017

  35. [43]

    Pytorch: An imperative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems , 32, 2019

    2019

  36. [44]

    Botorch: A framework for efficient monte-carlo bayesian optimization

    Maximilian Balandat, Brian Karrer, Daniel Jiang, Samuel Daulton, Ben Letham, Andrew G Wilson, and Eytan Bakshy. Botorch: A framework for efficient monte-carlo bayesian optimization. Advances in neural information processing systems , 33:21524--21538, 2020

    2020

  37. [45]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101 , 2017

    2017 arXiv

  38. [46]

    Random forests

    Leo Breiman. Random forests. Machine learning , 45(1):5--32, 2001

    2001

  39. [47]

    Transfer learning to model inertial confinement fusion experiments

    Kelli D Humbird, Jayson Luc Peterson, BK Spears, and Ryan G McClarren. Transfer learning to model inertial confinement fusion experiments. IEEE Transactions on Plasma Science , 48(1):61--70, 2019

    2019

  40. [48]

    Shock timing technique for the national ignition facility

    David H Munro, Peter M Celliers, Gilbert W Collins, David M Gold, Luiz B Da Silva, Steven W Haan, Robert C Cauble, Bruce A Hammel, and Warren W Hsing. Shock timing technique for the national ignition facility. Physics of Plasmas , 8(5):2245--2250, 2001

    2001

  41. [49]

    Machine learning for biological design

    Tom Blau, Iadine Chades, and Cheng Soon Ong. Machine learning for biological design. In Synthetic Biology: Methods and Protocols , pages 319--344. Springer, 2024

    2024

  42. [50]

    o ssische Technische Hochschule Z \

    Alkis Gotovos. Active learning for level set estimation. Master's thesis, Eidgen \"o ssische Technische Hochschule Z \"u rich, Department of Computer Science,, 2013

    2013

  43. [51]

    Unexpected improvements to expected improvement for bayesian optimization

    Sebastian Ament, Samuel Daulton, David Eriksson, Maximilian Balandat, and Eytan Bakshy. Unexpected improvements to expected improvement for bayesian optimization. Advances in Neural Information Processing Systems , 36:20577--20612, 2023

    2023

  44. [52]

    Combining multi-fidelity modelling and asynchronous batch bayesian optimization

    Jose Pablo Folch, Robert M Lee, Behrang Shafei, David Walz, Calvin Tsay, Mark van der Wilk, and Ruth Misener. Combining multi-fidelity modelling and asynchronous batch bayesian optimization. Computers & Chemical Engineering , 172:108194, 2023

    2023

  45. [53]

    mille-feuille: A Bayesian Optimisation wrapper for HPC scale simulations: https://github.com/aidancrilly/mille-feuille , 2025

    Aidan Crilly. mille-feuille: A Bayesian Optimisation wrapper for HPC scale simulations: https://github.com/aidancrilly/mille-feuille , 2025

    2025

  46. [54]

    The distribution of points in a cube and the accurate evaluation of integrals (in russian) zh

    I Sobol. The distribution of points in a cube and the accurate evaluation of integrals (in russian) zh. Vychisl. Mat. i Mater. Phys , 7:784--802, 1967

    1967

  47. [55]

    Progress toward a self-consistent set of 1d ignition capsule metrics in icf

    JD Lindl, SW Haan, OL Landen, AR Christopherson, and R Betti. Progress toward a self-consistent set of 1d ignition capsule metrics in icf. Physics of Plasmas , 25(12), 2018

    2018

  48. [56]

    S. P. Regan, R. Epstein, B. A. Hammel, L. J. Suter, J. Ralph, H. Scott, M. A. Barrios, D. K. Bradley, D. A. Callahan, C. Cerjan, G. W. Collins, S. N. Dixit, T. Doeppner, M. J. Edwards, D. R. Farley, S. Glenn, S. H. Glenzer, I. E. Golovkin, S. W. Haan, A. Hamza, D. G. Hicks, N....

    2012

  49. [57]

    Laser-plasma interactions and applications

    Paul McKenna, David Neely, Robert Bingham, and Dino Jaroszynski. Laser-plasma interactions and applications . Springer, 2013

    2013

  50. [58]

    Self-consistent growth rate of the rayleigh--taylor instability in an ablatively accelerating plasma

    Hideaki Takabe, K Mima, L Montierth, and RL Morse. Self-consistent growth rate of the rayleigh--taylor instability in an ablatively accelerating plasma. The Physics of fluids , 28(12):3676--3682, 1985

    1985

  51. [59]

    Growth rates of the ablative rayleigh--taylor instability in inertial confinement fusion

    R Betti, VN Goncharov, R \'a L McCrory, and C \'a P Verdon. Growth rates of the ablative rayleigh--taylor instability in inertial confinement fusion. Physics of Plasmas , 5(5):1446--1454, 1998

    1998

  52. [60]

    Betti, K

    R. Betti, K. Anderson, V. N. Goncharov, R. L. McCrory, D. D. Meyerhofer, S. Skupsky, and R. P. J. Town. Deceleration phase of inertial confinement fusion implosions. Physics of Plasmas , 9(5):2277--2286, 2002

    2002

  53. [61]

    Hohlraum reheating from burning nif implosions

    MS Rubery, MD Rosen, N Aybar, OL Landen, L Divol, CV Young, C Weber, J Hammer, JD Moody, AS Moore, et al. Hohlraum reheating from burning nif implosions. Physical Review Letters , 132(6):065104, 2024

    2024

  54. [62]

    Inertially confined fusion plasmas dominated by alpha-particle self-heating

    Omar A Hurricane, DA Callahan, DT Casey, EL Dewald, TR Dittrich, T D \"o ppner, S Haan, DE Hinkel, LF Berzak Hopkins, O Jones, et al. Inertially confined fusion plasmas dominated by alpha-particle self-heating. Nature Physics , 12(8):800--806, 2016

    2016

  55. [63]

    A 3d dynamic model to assess the impacts of low-mode asymmetry, aneurysms and mix-induced radiative loss on capsule performance across inertial confinement fusion platforms

    PT Springer, OA Hurricane, JH Hammer, R Betti, DA Callahan, EM Campbell, DT Casey, CJ Cerjan, D Cao, E Dewald, et al. A 3d dynamic model to assess the impacts of low-mode asymmetry, aneurysms and mix-induced radiative loss on capsule performance across inertial confinement fus...

    2018

  56. [64]

    Multidimensional modelling of Cross-Beam Energy Transfer for direct-drive Inertial Confinement Fusion

    Philip Moloney. Multidimensional modelling of Cross-Beam Energy Transfer for direct-drive Inertial Confinement Fusion . PhD thesis, Imperial College London, 2024

    2024

  57. [65]

    Progress in 2 mm glow discharge polymer mandrel development for nif

    A Nikroo, J Bousquet, R Cook, BW McQuillan, R Paguio, and M Takagi. Progress in 2 mm glow discharge polymer mandrel development for nif. Fusion science and technology , 45(2):165--170, 2004

    2004

  58. [66]

    Metrology feasibility study in support of the national direct-drive program

    H Huang, K Engelhorn, K Sequoia, A Greenwood, W Sweet, L Carlson, F Elsner, and M Farrell. Metrology feasibility study in support of the national direct-drive program. Fusion Science and Technology , 73(2):98--106, 2018

    2018

  59. [67]

    Impact of stalk on directly driven inertial confinement fusion implosions

    Maria Gatu Johnson, Patrick J Adrian, KS Anderson, BD Appelbe, JP Chittenden, AJ Crilly, D Edgell, CJ Forrest, Johan A Frenje, V Yu Glebov, et al. Impact of stalk on directly driven inertial confinement fusion implosions. Physics of Plasmas , 27(3), 2020

    2020

  60. [68]

    Exploring the parameter space of maglif implosions using similarity scaling

    DE Ruiz, PF Schmit, DA Yager-Elorriaga, MR Gomez, MR Weis, CA Jennings, AJ Harvey-Thompson, PF Knapp, SA Slutz, DJ Ampleford, et al. Exploring the parameter space of maglif implosions using similarity scaling. ii. current scaling. Physics of Plasmas , 30(3), 2023

    2023

  61. [69]

    Affordable, manageable, practical, and scalable (amps) high-yield and high-gain inertial fusion

    Andrew Alexander, Laura Robin Benedetti, Indrani Bhattacharyya, Jared Bowen, June Cabatu, Virgil Cacdac, Chhavi Chhavi, Chiatai Chen, Karen Chen, Dan Clark, et al. Affordable, manageable, practical, and scalable (amps) high-yield and high-gain inertial fusion. arXiv preprint a...

    2025 arXiv

  62. [70]

    A pushered capsule implosion as an alternate approach to the ignition regime for inertial confinement fusion

    SA MacLaren, DD-M Ho, OA Hurricane, EL Dewald, DA Martinez, RE Tipton, JE Pino, CV Young, HW Xu, CW Kong, et al. A pushered capsule implosion as an alternate approach to the ignition regime for inertial confinement fusion. Physics of Plasmas , 28(12), 2021

    2021

  63. [71]

    Energy gain of wetted-foam implosions with auxiliary heating for inertial fusion studies

    RW Paddock, TS Li, E Kim, JJ Lee, H Martin, RT Ruskov, S Hughes, SJ Rose, CD Murphy, RHH Scott, et al. Energy gain of wetted-foam implosions with auxiliary heating for inertial fusion studies. Plasma Physics and Controlled Fusion , 66(2):025005, 2023

    2023

  64. [72]

    A future of inertial confinement fusion without laser-plasma instabilities

    DH Froula, C Dorrer, A Cola \" tis, DH Edgell, RK Follett, EM Hill, IV Igumenshchev, AL Milder, JP Palastro, RC Shah, et al. A future of inertial confinement fusion without laser-plasma instabilities. Physics of Plasmas , 32(5), 2025

    2025

  65. [73]

    Rethinking optimization with differentiable simulation from a global perspective

    Rika Antonova, Jingyun Yang, Krishna Murthy Jatavallabhula, and Jeannette Bohg. Rethinking optimization with differentiable simulation from a global perspective. In Conference on robot learning , pages 276--286. PMLR, 2023

    2023

  66. [74]

    Generative neural reparameterization for differentiable pde-constrained optimization

    Archis S Joglekar. Generative neural reparameterization for differentiable pde-constrained optimization. arXiv preprint arXiv:2410.12683 , 2024

    2024 arXiv

  67. [75]

    Archer2 service description, December 2024

    George Beckett, Josephine Beech-Brandt, Kieran Leach, Zöe Payne, Alan Simpson, Lorna Smith, Andy Turner, and Anne Whiting. Archer2 service description, December 2024

    2024

  68. [76]

    Dropout injection at test time for post hoc uncertainty quantification in neural networks

    Emanuele Ledda, Giorgio Fumera, and Fabio Roli. Dropout injection at test time for post hoc uncertainty quantification in neural networks. Information Sciences , 645:119356, 2023

    2023

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.