Yihan Wang

Biography

Yihan Wang, PhD, is a Postdoctoral Researcher at ICSI and Lawrence Berkeley National Laboratory. Her research lies at the intersection of hydrology and scientific machine learning, with a focus on developing reliable, physically consistent, and interpretable models for hydrology and Earth system science.

Dr. Wang’s work advances modern deep learning architectures, including state-space models and diffusion-based generative models, for streamflow simulation and probabilistic forecasting. She also develops physics-aware neural networks that embed mass-conservation principles to investigate how physical constraints affect predictive performance, generalization, and learned representations across diverse hydroclimatic conditions.

At ICSI, Dr. Wang aims to advance deep learning methods that more effectively integrate hydrologic knowledge while fostering collaboration between the hydrologic science and artificial intelligence communities. Her broader goal is to develop accurate and reliable modeling tools for understanding and predicting water-system behavior, supporting hazard preparedness and informed water-resources decision-making.

Dr. Wang received her Ph.D. from the University of Oklahoma, her M.S. from the University of California, Los Angeles, and her B.E. from Hohai University.

Publications

Areas of Expertise

• Deep Learning and Neural Networks
• Generative AI and Foundation Models
• AI for Climate Change, Energy, and Environmental Sustainability
• GeoAI and Spatial Data Science Applications
• Reproducibility, Replicability, and Benchmarking in Research

ICSI Activity