Adaptive autoencoder latent space tuning for more robust machine learning beyond the training set for six-dimensional phase space diagnostics of a time-varying ultrafast electron-diffraction compact accelerator

Alexander Scheinker; Frederick Cropp; Daniele Filippetto

doi:10.1103/PhysRevE.107.045302

Adaptive autoencoder latent space tuning for more robust machine learning beyond the training set for six-dimensional phase space diagnostics of a time-varying ultrafast electron-diffraction compact accelerator

Phys Rev E. 2023 Apr;107(4-2):045302. doi: 10.1103/PhysRevE.107.045302.

Authors

Alexander Scheinker¹, Frederick Cropp^{2

3}, Daniele Filippetto²

Affiliations

¹ Applied Electrodynamics Group, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA.
² Lawrence Berkeley National Laboratory, One Cyclotron Road, Berkeley, California 94720, USA.
³ Department of Physics and Astronomy, University of California Los Angeles, Los Angeles, California 90095, USA.

PMID: 37198850
DOI: 10.1103/PhysRevE.107.045302

Abstract

We present a general adaptive latent space tuning approach for improving the robustness of machine learning tools with respect to time variation and distribution shift. We demonstrate our approach by developing an encoder-decoder convolutional neural network-based virtual 6D phase space diagnostic of charged particle beams in the HiRES ultrafast electron diffraction (UED) compact particle accelerator with uncertainty quantification. Our method utilizes model-independent adaptive feedback to tune a low-dimensional 2D latent space representation of ∼1 million dimensional objects which are the 15 unique 2D projections (x,y),...,(z,p_{z}) of the 6D phase space (x,y,z,p_{x},p_{y},p_{z}) of the charged particle beams. We demonstrate our method with numerical studies of short electron bunches utilizing experimentally measured UED input beam distributions.