The Jackson School of Geosciences is developing undergraduate classes in artificial intelligence as part of a broader push to incorporate artificial intelligence into the school.
Dapeng Feng and Fa Li, both hired as assistant professors in the Department of Earth and Planetary Sciences, use AI to study earth processes and they will teach courses in AI for undergraduates in the fall. Li will teach a class on the coding language Python and machine learning techniques, while Feng will be teaching an existing class on geoscience data.
“With our expertise, we can already do a lot of things,” Feng said. “(We want to) have our undergraduate students (realize) what amazing things AI can do in our field to provoke their interest for further study.”
Feng’s research focuses on hydrology and the water cycle with applications in flood and drought monitoring as well as water resource management. Li studies land environments and how they are affected by climate change.
“The problem (with) geosciences (is it’s) too big,” Li said. “We need to (combine) AI with domain knowledge together to develop (the) next generation (of) AI models.”
Both of their specialties use a “hybrid” model that combines physical models and machine learning to understand and study earth processes. Their research also represents a broader effort in geosciences to combine traditional physical modeling and AI machine learning to make sense of large amounts of data quickly.
“These digital platforms give us the leverage to inform decision makers and stakeholders and people that really need geoscience data,” said Danny Stockli, interim dean of the Jackson School of Geosciences. “They need it today, or maybe even yesterday, and not in 10 years.”
Feng and Li are also developing graduate-level courses that align with their research specialties.
“One of the things that is really important for us is the immediate translation of cutting-edge research to education,” Stockli said. “We already have a graduate stackable certificate in machine learning and data science, and so that just goes to show we’re not new to this field.”
The Jackson School is also emphasizing collaboration with the College of Natural Sciences and the Oden Institute for Computational Engineering & Sciences. The Jackson School and the Oden Institute are currently looking for a joint professor chair in computational geosciences, Stockli said.
In May, the University will also host HydroML, a symposium on machine learning and hydrology, which Feng is co-chairing.
“This is a great resource opportunity to UT, to the Jackson School, to Oden and also to our faculty and students,” Feng said. “We’re going to bring lots of expertise, lots of experts doing fantastic work on AI for machine learning and broader geoscience.”
