Our lab's mission is to design and develop accurate, scalable, and robust AI algorithms and tools that are inspired by unique challenges from interdisciplinary applications. Our vision is that future AI research requires the convergence of multiple disciplines, as real-world problems are so complex that one size AI does not fit all. We value both technical innovations in ML methodologies and tool deployment to solve a real problem that benefits society.
Current research topics include (but not limited to) spatiotemporal data mining, graph neural networks, self-supervised and weakly-supervised learning, physics-informed machine learning, robustness and uncertainty of AI models, large-scale distributed machine learning, as well as interdisciplinary applications in hydrology, disaster management, agriculture, transportation and smart cities, and public health. Our lab has established close collaborations with US Geological Survey, NOAA, NASA, NGA, Los Alamos National Lab, and industrial companies.