Chengyang Ying
Papers
1
Total Citations
90
H-Index
1
About
Chengyang Ying is a rising researcher at the forefront of physics-informed machine learning, a field that bridges data-driven AI with fundamental physical laws. His most-cited work, the 2022 survey "Physics-Informed Machine Learning: A Problems, Methods and Applications," has already garnered 90 citations, establishing itself as a key reference for researchers seeking to embed physical constraints into neural networks. Ying’s major contribution lies in systematically mapping how physical principles—such as conservation laws and differential equations—can guide machine learning models to produce more accurate, interpretable, and data-efficient solutions, particularly in scientific and engineering domains. This work addresses a critical gap: while deep learning excels in pattern recognition, it often fails to respect the underlying physics of real-world systems. By synthesizing diverse methodologies and applications, Ying has provided a roadmap for integrating prior physical knowledge into model design, training, and inference. His research is especially impactful for students and researchers working on problems where data is scarce or noisy, offering a principled approach to improving model robustness. As physics-informed machine learning continues to gain traction, Ying’s survey remains an essential starting point for anyone entering this transformative area.
Research Focus
Key Achievements
Top Papers
- 1