Invertible Surrogate Models: Joint surrogate modelling and reconstruction of Laser-Wakefield Acceleration by invertible neural networks

2021/06/01 by Bethke, Friedrich, Pausch, Richard, Stiller, Patrick +3 · 1 citation
#Accelerator Physics (physics.acc-ph) #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning (cs.LG) #Plasma Physics (physics.plasm-ph)

paper · doi:10.48550/arxiv.2106.00432

Abstract

Invertible neural networks are a recent technique in machine learning promising neural network architectures that can be run in forward and reverse mode. In this paper, we will be introducing invertible surrogate models that approximate complex forward simulation of the physics involved in laser plasma accelerators: iLWFA. The bijective design of the surrogate model also provides all means for reconstruction of experimentally acquired diagnostics. The quality of our invertible laser wakefield acceleration network will be verified on a large set of numerical LWFA simulations.

Cited by

• Surrogate modeling of capillary-discharge on-axis density profiles for laser guiding: forward prediction and inverse design

Related

• Surrogate modeling of capillary-discharge on-axis density profiles for laser guiding: forward prediction and inverse design
• Acceptance Rates of Invertible Neural Networks on Electron Spectra from Near-Critical Laser-Plasmas: A Comparison
• Surrogate Models studies for laser-plasma accelerator electron source design through numerical optimisation
• Improving Surrogate Model Accuracy for the LCLS-II Injector Frontend Using Convolutional Neural Networks and Transfer Learning
• Integration of Machine Learning-Based Plasma Acceleration Simulations into Geant4: A Case Study with the PALLAS Experiment