Yejiang Yang
Papers
3
Total Citations
30
H-Index
2
About
Yejiang Yang is a researcher advancing the intersection of neural network theory, robust optimization, and dynamical systems modeling. His work addresses fundamental challenges in ensuring reliability and scalability of neural models. In his highly cited 2022 paper, "Guaranteed approximation error estimation of neural networks and model modification" (24 citations), Yang introduced a rigorous framework for quantifying and correcting approximation errors in neural networks, providing critical theoretical guarantees for their deployment in safety-critical applications. Building on this, he developed a robust optimization framework using reachability analysis for training shallow neural networks (2021, 4 citations), where input data disturbances are modeled as interval sets to enhance noise resilience. Most recently, Yang proposed a distributed neural hybrid system learning framework (2024, 2 citations) that tackles scalability in modeling complex dynamical systems by mapping high-dimensional data to low-dimensional feature spaces. His contributions are particularly valuable for engineers and researchers working on trustworthy AI, control systems, and scientific machine learning, offering both theoretical foundations and practical methodologies for building more reliable and efficient neural network models.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3