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

Neural networks are being increasingly applied to control and decision-makingfor learning-enabled cyber-physical systems (LE-CPSs). They have shownpromising performance without requiring the development of complex physicalmodels; however, their adoption is significantly hindered by the concerns ontheir safety, robustness, and efficiency. In this work, we propose COCKTAIL, anovel design framework that automatically learns a neural network-basedcontroller from multiple existing control methods (experts) that could beeither model-based or neural network-based. In particular, COCKTAIL firstperforms reinforcement learning to learn an optimal system-level adaptivemixing strategy that incorporates the underlying experts withdynamically-assigned weights and then conducts a teacher-student distillationwith probabilistic adversarial training and regularization to synthesize astudent neural network controller with improved control robustness (measured bya safe control rate metric with respect to adversarial attacks or measurementnoises), control energy efficiency, and verifiability (measured by thecomputation time for verification). Experiments on three non-linear systemsdemonstrate significant advantages of our approach on these properties overvarious baseline methods.

Cocktail: Learn a Better Neural Network Controller from Multiple Experts via Adaptive Mixing and Robust Distillation