2021/03/08 by Yixuan Wang, Chao Huang, Wang, Yixuan +9 · 1 citation
Computer Science · Engineering · Physics and Astronomy · #Adversarial Robustness in Machine Learning #FOS: Electrical engineering #Fault Detection and Control Systems #Model Reduction and Neural Networks #Systems and Control (eess.SY) #cs.SY #eess.SY #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2103.05046
The paper has been accepted by Design Automation Conference 2021
arxiv created 2021/03/08 · openalex publication_date 2021/03/08 · arxiv updated 2021/03/10 · openalex created_date 2021/03/15 · openalex updated_date 2026/07/28
Neural networks are being increasingly applied to control and decision-making for learning-enabled cyber-physical systems (LE-CPSs). They have shown promising performance without requiring the development of complex physical models; however, their adoption is significantly hindered by the concerns on their safety, robustness, and efficiency. In this work, we propose COCKTAIL, a novel design framework that automatically learns a neural network-based controller from multiple existing control methods (experts) that could be either model-based or neural network-based. In particular, COCKTAIL first performs reinforcement learning to learn an optimal system-level adaptive mixing strategy that incorporates the underlying experts with dynamically-assigned weights and then conducts a teacher-student distillation with probabilistic adversarial training and regularization to synthesize a student neural network controller with improved control robustness (measured by a safe control rate metric with respect to adversarial attacks or measurement noises), control energy efficiency, and verifiability (measured by the computation time for verification). Experiments on three non-linear systems demonstrate significant advantages of our approach on these properties over various baseline methods.