Sanghani Center and CAIA cultivate transdisciplinary research in agriculture, AI, and data analytics
Photo of Anuj Karpatne, Sangwoo Kim and Venkat Sridhar standing together
A new initiative between the Center for Advanced Innovation in Agriculture (CAIA) and the Sanghani Center for Artificial Intelligence and Data Analytics has launched a Graduate Research Assistantship Program that provides scholarships for exceptional Ph.D. students conducting research that generates agricultural solutions enabled by artificial intelligence and data analytics.
The initiative builds upon a consistent collaboration between the two centers that began with the inauguration of CAIA in 2021. Faculty from the two centers have been working on joint research projects that includes a National Science Foundation-sponsored Convergence Accelerator project and some faculty are members of both centers.
“The collaboration underscores our shared dedication to harnessing data analytics and artificial intelligence to address the complex challenges facing the agricultural landscape,” said Brian Mayer, research scientist at the Sanghani Center who serves on CAIA’s executive committee.
Joint Graduate Scholarship
The Joint Graduate Scholarship supports Ph.D. students co-mentored by CAIA affiliate faculty and Sanghani Center faculty for up to two years.
“By providing resources to support graduate student co-mentorship, we are aiming to train truly transdisciplinary professionals,” said Robin White, associate professor in the School of Animal Sciences and associate director of CAIA. “Supporting collaboration and building capacity, the initiative will also help us generate new grant funding.”
To qualify for a scholarship, students must be enrolled in a graduate program at Virginia Tech with a focus on agriculture, engineering, data science, or related fields and propose research with relevance to both CAIA and Sanghani Center priorities, such as novel analysis of existing data sets, model-driven, or data-driven decision-making.
As part of the application process, students must also identify at least one faculty member from each center as co-advisors who will engage in formal co-mentorship, collaborate around their training and research activities, and commit to at least one more year of graduate research assistant funding for them.
“We are looking for applicants with a strong academic record who demonstrate an exceptional ability to learn and research potential,” Mayer said. “And for students who have a compelling vision of how the co-mentorship will create a transdisciplinary experience that will help them contribute to meaningful advancements at the nexus of agriculture, artificial intelligence, and data analytics.”
First scholarship recipients
Sangwoo Kim in the Department of Biological Systems Engineering and Runing Yang in the Bradley Department of Electrical and Computer Engineering are the first Ph.D. students to receive the Joint Graduate Scholarship.
Kim will integrate a sophisticated fluid dynamics deep learning model, the Environmental Fluid Dynamics Code, with Coupled Model Intercomparison Project Phase 6 data, employing a novel approach that can provide a comprehensive analysis on how the influx of sea and changing climate patterns affect rice production in the Mekong River Basin.
“I have dedicated my research to evaluating the effects of climate change on agricultural environments using physics-based hydrological models and remote sensing imagery, and recently developed an interest in statistical models such as machine learning and deep learning, which can overcome the limitations of physics-based models,” said Kim. “I believe that integrating these data-driven models with traditional physics-based models can overcome the limitations of existing agricultural research.”
His co-mentors for the project are Venkat Sridhar, associate professor in the Department of Biological Systems Engineering from CAIA, and Anuj Karpatne, associate professor in the Department of Computer Science from the Sanghani Center. Sridhar is an internationally recognized expert in assessing the impacts of climate change on hydrology and in water resources management. Karpatne is a pioneer in applying physics-based neural networks to ecosystems and aquatic systems.
“I look forward to receiving ongoing guidance and feedback to address complex scientific questions from my advisor Dr. Sridhar and by collaborating with Dr. Karpatne, I anticipate gaining profound insights into machine learning methods and enhancing my understanding of these technologies,” Kim said.
Photo of Haibo Huang, Runing Yang, and Ming Jin standing together
Yang is exploring how to leverage large language models’ (LLM) natural language processing capabilities to assist with meta-analysis in agriculture.
“Analyzing results from multiple studies will help reach more robust conclusions and reveal subtle discoveries,” said Yang.
He will develop a framework to interpret results and generate data-driven hypotheses that can answer impactful research questions, such as why zero-tillage — meaning land where crops are grown is not tilled between harvest and sowing — might yield more than conventional tillage, where most of the crop residue is incorporated or buried into the soil.
“In agriculture, where data is precious and experiments span years, quality meta-analysis that enables synthesis is valuable but very time-consuming. The ability to process text will significantly save manual labor and accelerate data processing and synthesis across literature pertaining to this field of scientific research,” Yang said.
His co-mentors are Ming Jin, assistant professor in the Bradley Department of Electrical and Computer Engineering, from the Sanghani Center, and from CAIA, Haibo Huang, associate professor in the Department of Food Science and Technology and affiliate faculty in the Department of Biological Systems Engineering.
Jin’s expertise is in trustworthy machine learning with focus on safe reinforcement learning and foundation models. Huang’s research focuses on developing food- and bio-processing technologies to produce food ingredients, animal feed, and green chemicals from agricultural and food products for improving the sustainability of the food production chain.
“I believe their co-mentorship ensures that the project is technically sound and meaningful for real-world agricultural research,” Yang said. “My advisor Dr. Jin will continue to guide me with the latest LLM methods suitable for agricultural data processing and on frameworks that improve causal inference, and Dr. Huang will provide me with background knowledge and insights into which features or experiment settings significantly impact conclusions; assist with interpreting results; and help ensure the verified hypotheses are valid and applicable.”
Kim and Yang were chosen by a panel of representatives from both CAIA and the Sanghani Center. All proposals were reviewed based on their quality, feasibility, innovation, and alignment with the program's objectives.
