- Physical Review Journals
All JournalsPhysics Magazine

Physical Review Research

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access

High-dimensional inverse design of inertial fusion implosions via differentiable simulation

A. J. Crilly1,2,*, P. Travis3, J. P. Brodrick4, J. B. Coughlin3, and A. S. Joglekar3,4

  • 1Centre for Inertial Fusion Studies, The Blackett Laboratory, Imperial College, London SW7 2AZ, United Kingdom
  • 2I-X Centre for AI In Science, Imperial College London, White City Campus, 84 Wood Lane, London W12 0BZ, United Kingdom
  • 3Ergodic LLC, Seattle, Washington 98103, USA
  • 4Pasteur Labs, Brooklyn, New York 11205, USA

  • *Contact author: ac116@ic.ac.uk
PDF

Phys. Rev. Research 8, 033353 – Published 23 September, 2026

DOI: https://doi.org/10.1103/bpms-63ml

Abstract

Inertial confinement fusion implosion design requires simultaneous optimization of strongly coupled target and driver parameters across high-dimensional design spaces. Existing automated design approaches typically rely on nondifferentiable radiation-hydrodynamics codes treated as black boxes, making optimization increasingly expensive as dimensionality grows. In this work, we present a differentiable simulation approach for high-dimensional inverse design of inertial confinement fusion implosions. Automatic differentiation through a differentiable implosion physics model, driven by an external pressure pulse, provides gradients of implosion objectives with respect to design parameters, enabling gradient-based optimization. The framework is applied to 25 kJ OMEGA-scale direct-drive implosions, optimizing 500-parameter laser pulses across sampled target geometries. The optimized pulse recovers a near-isentropic rise to peak power without that structure (i.e., pulse shape) being imposed. Neural-network pulse parametrizations are additionally explored as a means of accelerating design-space exploration. These results establish differentiable implosion modeling as a promising tool for inertial confinement fusion design, while motivating further work on adjoint robustness and higher-fidelity differentiable simulators.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (64)

  1. S. Atzeni and J. Meyer-ter Vehn, The Physics of Inertial Fusion: BeamPlasma Interaction, Hydrodynamics, Hot Dense Matter (Oxford University Press, Oxford, 2004).
  2. J. Lindl, Development of the indirect-drive approach to inertial confinement fusion and the target physics basis for ignition and gain, Phys. Plasmas 2, 3933 (1995).
  3. R. S. Craxton, K. S. Anderson, T. R. Boehly, V. N. Goncharov, D. R. Harding, J. P. Knauer, R. L. McCrory, P. W. McKenty, D. D. Meyerhofer, J. F. Myatt, et al., Direct-drive inertial confinement fusion: A review, Phys. Plasmas 22, 110501 (2015).
  4. H. Abu-Shawareb, R. Acree, P. Adams, J. Adams, B. Addis, R. Aden, P. Adrian, B. B. Afeyan, M. Aggleton, L. Aghaian, et al., Lawson criterion for ignition exceeded in an inertial fusion experiment, Phys. Rev. Lett. 129, 075001 (2022).
  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, et al., Experimental achievement and signatures of ignition at the National Ignition Facility, Phys. Rev. E 106, 025202 (2022).
  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, et al., Design of an inertial fusion experiment exceeding the Lawson criterion for ignition, Phys. Rev. E 106, 025201 (2022).
  7. V. Gopalaswamy, A. Lees, R. Ejaz, C. A. Thomas, T. J. B. Collins, K. S. Anderson, W. Ebmeyer, and R. Betti, Automated and highly parallelized Bayesian optimization scheme for direct drive fusion experiments on Omega, Phys. Rev. Res. 7, 013009 (2025).
  8. P. W. Hatfield, S. J. Rose, and R. H. H. Scott, The blind implosion-maker: Automated inertial confinement fusion experiment design, Phys. Plasmas 26, 062706 (2019).
  9. V. Gopalaswamy, R. Betti, J. P. Knauer, A. Lees, D. Patel, A. R. Christopherson, I. V. Igumenshchev, D. Cao, K. S. Anderson, A. Shvydky, et al., Using statistical modeling to predict and understand fusion experiments, Phys. Plasmas 28, 122705 (2021).
  10. R. Ejaz, V. Gopalaswamy, A. Lees, C. Kanan, D. Cao, and R. Betti, Deep learning-based predictive models for laser direct drive at the Omega Laser Facility, Phys. Plasmas 31, 052703 (2024).
  11. M. Brönner, S. Atzeni, D. Callahan, J. Gaffney, P. Gibbon, L. C. Jarrott, A. Mateo, L. Savino, N. Schott, W. Theobald, et al., Particle swarm optimization of 1D isochoric compression designs for fast ignition, Phys. Plasmas 32, 022710 (2025).
  12. J. Wang, N. Chiang, A. Gillette, and J. L. Peterson, A multifidelity Bayesian optimization method for inertial confinement fusion design, Phys. Plasmas 31, 032706 (2024).
  13. B. K. Spears, S. Brandon, D. T. Casey, J. E. Field, J. A. Gaffney, K. D. Humbird, A. L. Kritcher, M. K. G. Kruse, E. Kur, B. Kustowski, et al., Predicting fusion ignition at the National Ignition Facility with physics-informed deep learning, Science 389, 727 (2025).
  14. K. D. Humbird, J. L. Peterson, B. K. Spears, and R. G. McClarren, Transfer learning to model inertial confinement fusion experiments, IEEE Trans. Plasma Sci. 48, 61 (2019).
  15. N. N. Vazirani, M. J. Grosskopf, D. J. Stark, P. A. Bradley, B. M. Haines, E. Loomis, S. L. England, and W. A. Scales, Coupling 1D xRAGE simulations with machine learning for graded inner shell design optimization in double shell capsules, Phys. Plasmas 28, 122709 (2021).
  16. N. N. Vazirani, R. Sacks, B. M. Haines, M. J. Grosskopf, D. J. Stark, and P. A. Bradley, Bayesian batch optimization for molybdenum versus tungsten inertial confinement fusion double shell target design, Stat. Anal. Data Min. 17, e11698 (2024).
  17. N. N. Vazirani, R. F. Sacks, B. M. Haines, M. J. Grosskopf, D. J. Stark, P. A. Bradley, E. N. Loomis, E. C. Merritt, and H. F. Robey, Comparison of tungsten versus molybdenum for double shell capsules using machine learning design optimization, Phys. Plasmas 32, 032706 (2025).
  18. J. Bradbury, R. Frostig, P. Hawkins, M. J. Johnson, Y. Katariya, C. Leary, D. Maclaurin, G. Necula, A. Paszke, J. VanderPlas, S. Wanderman-Milne, and Q. Zhang, JAX: Composable transformations of Python + NumPy programs (2018), http://github.com/jax-ml/jax.
  19. T. R. Boehly, D. L. Brown, R. S. Craxton, R. L. Keck, J. P. Knauer, J. H. Kelly, T. J. Kessler, S. A. Kumpan, S. J. Loucks, S. A. Letzring, et al., Initial performance results of the OMEGA laser system, Opt. Commun. 133, 495 (1997).
  20. J. L. Lions, Optimal Control of Systems Governed by Partial Differential Equations (Springer-Verlag, Berlin, 1971), Vol. 170.
  21. P. Kidger, On neural differential equations, arXiv:2202.02435.
  22. M. Giles and P. Glasserman, Smoking adjoints: Fast Monte Carlo Greeks, Risk 19, 88 (2006).
  23. A. Gholami, K. Keutzer, and G. Biros, ANODE: Unconditionally accurate memory-efficient gradients for neural ODEs, arXiv:1902.10298.
  24. A. S. Joglekar, Generative neural reparameterization for differentiable PDE-constrained optimization, arXiv:2410.12683.
  25. A. S. Joglekar and A. G. R. Thomas, Unsupervised discovery of nonlinear plasma physics using differentiable kinetic simulations, J. Plasma Phys. 88, 905880608 (2022).
  26. A. S. Joglekar and A. G. R. Thomas, Machine learning of hidden variables in multiscale fluid simulation, Mach. Learn.: Sci. Technol. 4, 035049 (2023).
  27. K. G. Miller, T. E. Gutierrez, A. S. Joglekar, A. Elliott, D. H. Froula, and J. P. Palastro, Spatiotemporal control of laser intensity using differentiable programming, Mach. Learn.: Sci. Technol. 7, 025014 (2026).
  28. A. L. Milder, A. S. Joglekar, W. Rozmus, and D. H. Froula, Qualitative and quantitative enhancement of parameter estimation for model-based diagnostics using automatic differentiation with an application to inertial fusion, Mach. Learn.: Sci. Technol. 5, 015026 (2024).
  29. J. Citrin, I. Goodfellow, A. Raju, J. Chen, J. Degrave, C. Donner, F. Felici, P. Hamel, A. Huber, D. Nikulin, et al., TORAX: A fast and differentiable tokamak transport simulator in JAX, arXiv:2406.06718.
  30. N. McGreivy, S. R. Hudson, and C. Zhu, Optimized finite-build stellarator coils using automatic differentiation, Nucl. Fusion 61, 026020 (2021).
  31. D. Panici, R. Conlin, D. W. Dudt, K. Unalmis, and E. Kolemen, The DESC stellarator code suite. Part 1. Quick and accurate equilibria computations, J. Plasma Phys. 89, 955890303 (2023).
  32. R. Conlin, D. W. Dudt, D. Panici, and E. Kolemen, The DESC stellarator code suite. Part 2. Perturbation and continuation methods, J. Plasma Phys. 89, 955890305 (2023).
  33. P. Kidger and C. Garcia, equinox: Neural networks in JAX via callable PyTrees and filtered transformations, in Differentiable Programming Workshop at Neural Information Processing Systems (NeurIPS, 2021).
  34. M. C. Herrmann, M. Tabak, and J. D. Lindl, A generalized scaling law for the ignition energy of inertial confinement fusion capsules, Nucl. Fusion 41, 99 (2001).
  35. J. D. Lindl, S. W. Haan, O. L. Landen, A. R. Christopherson, and R. Betti, Progress toward a self-consistent set of 1D ignition capsule metrics in ICF, Phys. Plasmas 25, 122704 (2018).
  36. J. Von Neumann and R. D. Richtmyer, A method for the numerical calculation of hydrodynamic shocks, J. Appl. Phys. 21, 232 (1950).
  37. A. J. Kemp and J. Meyer-ter Vehn, An equation of state code for hot dense matter, based on the QEOS description, Nucl. Instrum. Methods Phys. Res. Sect. A 415, 674 (1998).
  38. R. M. More, K. H. Warren, D. A. Young, and G. B. Zimmerman, A new quotidian equation of state (QEOS) for hot dense matter, Phys. Fluids 31, 3059 (1988).
  39. E. M. Epperlein and M. G. Haines, Plasma transport coefficients in a magnetic field by direct numerical solution of the Fokker–Planck equation, Phys. Fluids 29, 1029 (1986).
  40. J. D. Huba, NRL Plasma Formulary (Naval Research Laboratory, 1998), Vol. 6790.
  41. K. Korner, B. Talamini, J. Andrej, M. Tupek, W. Moses, D. Tortorelli, R. Rieben, T. Kolev, J. Bramwell, D. White, et al., Differentiable Lagrangian shock hydrodynamics with application to stable shock acceleration of density interfaces, Comput. Methods Appl. Mech. Eng. 451, 118663 (2026).
  42. J. Thrussell and J. M. Ferguson, ExactPack: A Python library of exact analytic solutions, SoftwareX 24, 101560 (2023).
  43. R. Betti, A. R. Christopherson, B. K. Spears, R. Nora, A. Bose, J. Howard, K. M. Woo, M. J. Edwards, and J. Sanz, Alpha heating and burning plasmas in inertial confinement fusion, Phys. Rev. Lett. 114, 255003 (2015).
  44. A. R. Christopherson, R. Betti, S. Miller, V. Gopalaswamy, O. M. Mannion, and D. Cao, Theory of ignition and burn propagation in inertial fusion implosions, Phys. Plasmas 27, 052708 (2020).
  45. I. Sobol, The distribution of points in a cube and the accurate evaluation of integrals, U.S.S.R. Comput. Math. Math. Phys. 7, 86 (1967).
  46. A. J. Crilly, P. W. Moloney, D. Shi, and E. A. Ferdinandi, Automated simulation-based design via multi-fidelity active learning and optimization for laser direct drive implosions, Phys. Plasmas 33, 012702 (2026).
  47. J. Nuckolls, L. Wood, A. Thiessen, and G. Zimmerman, Laser compression of matter to super-high densities: Thermonuclear (CTR) applications, Nature (London) 239, 139 (1972).
  48. S. Hoyer, J. Sohl-Dickstein, and S. Greydanus, Neural reparameterization improves structural optimization, arXiv:1909.04240.
  49. M. Tancik, P. Srinivasan, B. Mildenhall, S. Fridovich-Keil, N. Raghavan, U. Singhal, R. Ramamoorthi, J. Barron, and R. Ng, Fourier features let networks learn high frequency functions in low dimensional domains, Adv. Neural Inf. Process. Syst. 33, 7537 (2020).
  50. S. Wang, H. Wang, and P. Perdikaris, On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks, Comput. Methods Appl. Mech. Eng. 384, 113938 (2021).
  51. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, Attention is all you need, arXiv:1706.03762.
  52. B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, NeRF: Representing scenes as neural radiance fields for view synthesis, arXiv:2003.08934.
  53. M. Telgarsky, Benefits of depth in neural networks, arXiv:1602.04485.
  54. D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, arXiv:1412.6980.
  55. J. Snoek, H. Larochelle, and R. P Adams, Practical Bayesian optimization of machine learning algorithms, Adv. Neural Inf. Process. Syst. 25, 2951 (2012).
  56. B. Sanderse, P. Stinis, R. Maulik, and S. E Ahmed, Scientific machine learning for closure models in multiscale problems: A review, arXiv:2403.02913.
  57. K. Um, R. Brand, Y. R. Fei, P. Holl, and N. Thuerey, Solver-in-the-loop: Learning from differentiable physics to interact with iterative PDE-solvers, Adv. Neural Inf. Process. Syst. 33, 6111 (2020).
  58. N. Bouziani, D. A Ham, and A. Farsi, Differentiable programming across the PDE and machine learning barrier, arXiv:2409.06085.
  59. A. J. Crilly, B. Duhig, and N. Bouziani, Learning closure relations using differentiable programming: An example in radiation transport, J. Quant. Spectrosc. Radiat. Transfer 318, 108941 (2024).
  60. V. Gopalaswamy, R. Betti, J. P. Knauer, N. Luciani, D. Patel, K. M. Woo, A. Bose, I. V. Igumenshchev, E. M. Campbell, K. S. Anderson, et al., Tripled yield in direct-drive laser fusion through statistical modelling, Nature (London) 565, 581 (2019).
  61. R. K. Follett, A. Colaïtis, I. V. Igumenshchev, D. Cao, R. Nora, V. N. Goncharov, D. N. Polsin, R. Betti, J. P. Palastro, T. J. B. Collins, et al., An experimentally informed design process for future inertial confinement fusion facilities, Phys. Plasmas 32, 042705 (2025).
  62. J. Delettrez, R. Epstein, M. C. Richardson, P. A. Jaanimagi, and B. L. Henke, Effect of laser illumination nonuniformity on the analysis of time-resolved X-ray measurements in UV spherical transport experiments, Phys. Rev. A 36, 3926 (1987).
  63. A. Lees, R. Betti, V. Gopalaswamy, J. P. Knauer, L. A. Ceurvorst, D. Patel, R. Ejaz, K. S. Anderson, K. A. Bauer, M. J. Bonino, et al., Approaching hydro-equivalent ignition in laser direct-drive via target design optimization using novel statistical modeling, Phys. Plasmas 32, 072707 (2025).
  64. C. Rackauckas, The numerical analysis of differentiable simulation: How automatic differentiation of physics can give incorrect derivatives, in ISCL Seminar Series (2025).

Outline

Information

Phys. Rev. Research 8, 033353– Published 23 September, 2026
  • Received 18 June 2026
  • Accepted 1 September 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation