Research papers and current work.

Audited Surrogate Gradients for Efficient Model-Based Flow Control

Ruige Kong, Stefania Fresca

Audits surrogate action gradients against high-fidelity finite differences, extracts a compact feedback law, and anchors policy optimization to that law before validating controllers in the original solver.

Audited Surrogate Gradients for Efficient Model-Based Flow Control

Ruige Kong, Stefania Fresca

Audits surrogate action gradients against high-fidelity finite differences, extracts a compact feedback law, and anchors policy optimization to that law before validating controllers in the original solver.

Spatial Context or Memory? Sizing the Inputs of Neural Surrogates for Partially Observed Fields

Ruige Kong

Studies when temporal history can replace spatial context in forecasting from sparse or noisy sensors, using exact Gaussian conditioning and learned models of linear fields, Burgers flow, and the Kuramoto–Sivashinsky equation.

Comparative and Interpretative Analysis of CNN and Transformer Models in Predicting Wildfire Spread Using Remote Sensing Data

Yihang Zhou, Ruige Kong, Zhengsen Xu, Linlin Xu, Sibo Cheng

Compares CNN and transformer architectures for next day wildfire segmentation under severe class imbalance. SHAP, Grad-CAM, and Integrated Gradients characterize architecture-dependent feature attribution.

DOI: 10.1029/2024JH000409

@article{zhou2025wildfire,
  title={Comparative and Interpretative Analysis of CNN and
         Transformer Models in Predicting Wildfire Spread
         Using Remote Sensing Data},
  author={Zhou, Yihang and Kong, Ruige and Xu, Zhengsen and
          Xu, Linlin and Cheng, Sibo},
  journal={Journal of Geophysical Research:
           Machine Learning and Computation},
  volume={2},
  number={2},
  year={2025},
  doi={10.1029/2024JH000409}
}