Laser Wakefield Accelerator modelling with Variational Neural Networks

M. J.V. Streeter*, C. Colgan, C. C. Cobo, C. Arran, E. E. Los, R. Watt, N. Bourgeois, L. Calvin, J. Carderelli, N. Cavanagh, S. J.D. Dann, R. Fitzgarrald, E. Gerstmayr, A. S. Joglekar, B. Kettle, P. Mckenna, C. D. Murphy, Z. Najmudin, P. Parsons, Q. QianP. P. Rajeev, C. P. Ridgers, D. R. Symes, A. G.R. Thomas, G. Sarri, S. P.D. Mangles

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


A machine learning model was created to predict the electron spectrum generated by a GeVclass laser wakefield accelerator. The model was constructed from variational convolutional neural networks which mapped the results of secondary laser and plasma diagnostics to the generated electron spectrum. An ensemble of trained networks was used to predict the electron spectrum and to provide an estimation of the uncertainty on that prediction. It is anticipated that this approach will be useful for inferring the electron spectrum prior undergoing any process which can alter or destroy the beam. In addition, the model provides insight into the scaling of electron beam properties due to stochastic fluctuations in the laser energy and plasma electron density.

Original languageEnglish
JournalHigh Power Laser Science and Engineering
Publication statusAccepted/In press - 6 Jan 2023

Bibliographical note

© The Author(s), 2023

Cite this