physics.acc-phOct 27, 2026

Inverse modelling of a laser-plasma accelerator

Authors: Mykyta Lenivenko, Arnaud Beck, Hayg Guler, Gueladio Kane, Sophie Kazamias, Viacheslav Kubytskyi, Bruno Lucas, Marceau Pintonato, 

Organizations: IJCLab · LLR

Abstract

The electron beams of a laser-plasma accelerator depend non-linearly on the laser and gas settings. This dependence is explored using particle-in-cell (PIC) simulations, and machine-learning models trained on these simulations predict the beam from the settings. Here we address the reverse problem: given a measured beam, which settings produced it? We infer three of them: the gas pressure p_1p\_1, the normalised vector potential a_0a\_0 and the nitrogen concentration c_N_2c\_{N\_2}. The solution is not guaranteed to be unique. We train an ensemble of mixture density networks that, for each beam, output several possible sets of settings with their probabilities and uncertainties, and test them on simulated beams with realistic measurement noise. On average, the predictions are accurate (R2=0.91R^2 = 0.91), yet for about two-thirds of the beams the inferred distribution has two competing solutions. These share nearly the same value of a_02p_1a\_0^2 p\_1, which controls how strongly the laser focuses itself in the plasma and, in earlier PALLAS studies, when electrons start to be injected. Keeping this quantity fixed gives beams that are practically indistinguishable with realistic diagnostics. A forward model used as a check removes unphysical solutions but cannot choose between the two plausible ones. We then test two ways to resolve the ambiguity. Adding the full electron energy spectrum improves the mean prediction (R2=0.94R^2 = 0.94), mostly through a_0a\_0. It changes the weights of the two solutions but does not move them. A second measurement at a deliberately different plasma density is more effective. In paired PIC simulations, it reduces the error on a_0a\_0 by a factor of 2.6, whereas repeating the measurement at the same density brings almost no gain.

Explore similar work

CardsList
  1. Energy-based models for diagnostic reconstruction and analysis in a laboratory plasma device

    May 9, 2026Phil Travis, Troy CarterEnergy-Based ModelsConditional Generative Modeling

  2. Learning a Latent Pulse Shape Interface for Photoinjector Laser Systems

    Feb 19, 2026Alexander Klemps, Denis Ilia, Pradeep Kr. Banerjee +3Representation LearningGenerative Modeling

  3. AI4Plasma I: Data-Efficient Surrogate Modeling of Capacitively Coupled Plasma Fields

    Nov 11, 2026Zhenyu Zou, Junjue Liao, Hao Yang +11Neural Surrogate ModelingSurrogate Modeling