🎉 Accepted by IEEE TAC as a full paper (https://t.co/pPaiYdAmm8)! Our work achieves certified safe control from offline data, without system identification, while guaranteeing safety, robustness, and recursive feasibility. #ControlTheory#DataDrivenControl#AutonomousDriving
@JDVance Please respect your presidents, wether you like them or not. They are an integral part of this nation's history. Unlike countries with deep historical roots, the U.S. could learn to better appreciate and honor its past. Making fun of former president is naive and disrespectful.
Two papers accepted by ICMLA'24: 1) Approximate Bisimulation Relation Restoration for Neural Networks Based On Knowledge Distillation; 2) Discrepancy-Based Knowledge Distillation for Image Classification Restoration.
Two papers accepted by MECC 2024: 1) Data-Driven Decentralized Control Design for Discrete-Time Large-Scale Systems; 2) Efficient Neural Hybrid System Learning and Interpretable Transition System Abstraction for Dynamical Systems (joint LDSC).
Two papers accepted by ACC 2024. (1) A Transition System Abstraction Framework for Neural Network Dynamical System Models; (2) Compression Repair for Feedforward Neural Networks Based on Model Equivalence Evaluation.
Our new paper accepted by the International Journal of Control, available online: Observer-based safety monitoring of nonlinear dynamical systems with neural networks via quadratic constraint approach https://t.co/PgYLVRzMZN
“Computationally efficient neural hybrid automaton framework for learning complex dynamics.” A novel modeling method using shallow neural networks and hybrid system architecture to build computationally efficient and highly accurate data-driven model. https://t.co/yUgh2kkfwu
Two papers "A Data-Driven Hybrid Automaton Framework to Modeling Complex Dynamical Systems" and "Guaranteed Quantization Error Computation for Neural Network Model Compression" accepted by ICIT 2023. Notably, Wesley Cooke, an undergraduate student, is leading the second paper.
Augusta University celebrates @AUG_Cyber's first Master of Science in Computer Science graduate, Patrick Woolard.
Congrats on being an #AUGGRAD#LikeNoOther, Patrick! https://t.co/uJY85ZWnTq
Our @NSF grant on human cyber-physical systems (h-CPS) got awarded! I will be collaborating with @jeoresearch studying data-driven modeling and control of h-CPS with VR interfaces.
I am looking for highly self-motivated PhD students with interests in one or more of the following areas to join my group: control systems, machine learning, formal methods, and cyber-physical systems. PhD Student Opening details: https://t.co/TDBXwWuN6O