English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning (ML) tools such as encoder-decoder deep convolutional neuralnetworks (CNN) are able to extract relationships between inputs and outputs oflarge complex systems directly from raw data. For time-varying systems thepredictive capabilities of ML tools degrade as the systems are no longeraccurately represented by the data sets with which the ML models were trained.Re-training is possible, but only if the changes are slow and if newinput-output training data measurements can be made online non-invasively. Inthis work we present an approach to deep learning for time-varying systems inwhich adaptive feedback based only on available system output measurements isapplied to encoded low-dimensional dense layers of encoder-decoder type CNNs.We demonstrate our method in developing an inverse model of a complex chargedparticle accelerator system, mapping output beam measurements to input beamdistributions while both the accelerator components and the unknown input beamdistribution quickly vary with time. We demonstrate our results usingexperimental measurements of the input and output beam distributions of theHiRES ultra-fast electron diffraction (UED) microscopy beam line at LawrenceBerkeley National Laboratory. We show how our method can be used to aid bothphysics and ML-based surrogate online models to provide non-invasive beamdiagnostics and we also demonstrate how our method can be used to automaticallytrack the time varying quantum efficiency map of a particle accelerator'sphotocathode.

Adaptive deep learning for time-varying systems with hidden parameters: Predicting changing input beam distributions of compact particle accelerators