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

Recovering a globally accurate complex physics field from limited sensor iscritical to the measurement and control in the aerospace engineering. Generalreconstruction methods for recovering the field, especially the deep learningwith more parameters and better representational ability, usually require largeamounts of labeled data which is unaffordable. To solve the problem, this paperproposes Uncertainty Guided Ensemble Self-Training (UGE-ST), using plentifulunlabeled data to improve reconstruction performance. A novel self-trainingframework with the ensemble teacher and pretraining student designed to improvethe accuracy of the pseudo-label and remedy the impact of noise is firstproposed. On the other hand, uncertainty-guided learning is proposed toencourage the model to focus on the highly confident regions of pseudo-labelsand mitigate the effects of wrong pseudo-labeling in self-training, improvingthe performance of the reconstruction model. Experiments include the pressurevelocity field reconstruction of airfoil and the temperature fieldreconstruction of aircraft system indicate that our UGE-ST can save up to 90%of the data with the same accuracy as supervised learning.

Uncertainty Guided Ensemble Self-Training for Semi-Supervised Global Field Reconstruction