Differentiable control of laser–plasma experiments with conditional diffusion models
Machine Learning: Science and Technology · 01.12.2026 · 7 (6) · 065001
DOI: 10.1088/2632-2153/aeb79a ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning: Science and Technology, online erschienen am 30.10.2026. Verfasst von Matěj Mráček und 3 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Grantová Agentura České Republiky, National Science Foundation.
Abstract (Originalsprache)
Tuning a laser–wakefield accelerator to a desired electron energy spectrum is normally done by scanning the machine one setting at a time, which consumes scarce beam time and expert attention. We address the inverse problem directly, recovering the experimental settings that produce a target spectrum. To do so we train a conditional diffusion model, built on the Elucidating the Design Space of Diffusion-Based Models (EDM) framework, on measured electron spectra from the ALFA beamline, conditioned on laser energy, gas-jet backing pressure, and gas-valve opening time. Because EDM samples by integrating a probability-flow ordinary differential equation, the map from settings to spectrum is differentiable end to end, and the settings themselves can be treated as free variables and optimized by backpropagation through the sampler. We propose and evaluate a three-stage strategy, BO → L-BFGS → RAdam, in which Bayesian optimization provides a global warm start across flat regions, L-BFGS exploits local curvature, and a final stochastic phase relaxes the solution away from sharp, bin-aligned minima. Against seven alternative single- and multi-stage strategies, this ordering recovers the target settings with the lowest mean squared error. We position this as a tool for differentiable inverse design in experiment planning, complementary to particle-in-cell simulation rather than a substitute for it.
Förderung
- Grantová Agentura České Republiky
- National Science Foundation
Zitieren
APA
Mráček, M., Grittani, G. M., Lazzarini, C. M., & Kovalenko, A. (2026). Differentiable control of laser–plasma experiments with conditional diffusion models. Machine Learning: Science and Technology, 7(6), 065001. https://doi.org/10.1088/2632-2153/aeb79a
Vancouver
Mráček M, Grittani GM, Lazzarini CM, Kovalenko A. Differentiable control of laser–plasma experiments with conditional diffusion models. Machine Learning: Science and Technology. 2026;7(6):065001. doi:10.1088/2632-2153/aeb79a
BibTeX
@article{Mrek2026,
author = {Mráček, Matěj and Grittani, Gabriele Maria and Lazzarini, Carlo Maria and Kovalenko, Alexander},
title = {Differentiable control of laser–plasma experiments with conditional diffusion models},
journal = {Machine Learning: Science and Technology},
year = {2026},
volume = {7},
number = {6},
pages = {065001},
doi = {10.1088/2632-2153/aeb79a}
}
EUVIMED