Researchers are developing an artificial intelligence tool that uses snapshots to help farmers identify pests and protect crops. Enlarged image. Photo montage courtesy of Aditya Balu.
AMES, Iowa – Iowa State University researchers are developing a large-scale, vision-based artificial intelligence tool to identify agricultural pests and ultimately recommend control, and are launching a new National Artificial Intelligence Research Resource Pilot was one of the first teams to obtain computing support from
The award provides Iowa State University researchers with 1 million “node hours” of supercomputing time on the $60 million Frontera supercomputer at the Texas Advanced Computing Center at the University of Texas at Austin . Frontera's power comes from more than 8,000 compute nodes, each containing 56 core processors. It is the fastest supercomputer based on a university campus in the United States.
Using Frontera, Iowa State researchers will be able to train a large ensemble of machine learning models that can analyze photos to quickly identify agricultural pests such as insects and weeds. The model is packaged into an app platform and designed to be deployed globally to help farmers protect their crops.
Future work will include connecting the vision model with a coordinated large-scale language model to provide a conversational tool that can suggest pest control strategies.
Bhaskar Ganapathy Subramanian, the Joseph and Elizabeth Anderlik Professor of Engineering at Iowa State University, is leading the computing project with Aditya Bal, a data scientist at the Iowa State Center for Translational AI . (See sidebar for a complete list of team members.)
Ganapathysubramanian said the center's computing support will improve the research team's ability to train, deploy and evaluate sophisticated AI models that enable profitable and sustainable agriculture.
He gave a presentation on the project today at the White House Office of Science and Technology Policy's Opportunities in AI Research Frontiers launch event in Washington, DC.
During the event, the National Intelligence and Research Resources Pilot, established in January as a two-year pilot program led by the National Science Foundation (NSF), announced that it will support an initial 35 projects by providing access to advanced computers. Announced.
NSF Director Sethuraman Panchanathan said the pilot is “driven by the need to advance responsible AI research and expand access to cutting-edge resources needed for AI research.”
Ganapathy Subramanian of Iowa State University said the researchers have been working on this approach for about two years, with the support of two federally funded efforts on campus. That is AIIRA, the AI Institute for Resilient Agriculture. COALESCE, Context Aware Learning for Sustainable Cyber Agriculture Systems.
The original idea conceptualized by the Iowa Soinomics team, a research team founded in 2014 that applies technology and data science to agricultural improvement, was to build an AI tool to identify diseases in soybeans.
Ganapathy Subramanian said the idea evolved to include identifying insects harmful to soybeans, then agricultural pests across Iowa and then around the world.
“The range is gradually expanding,” he said. “And so are the computing requirements for developing our models.”
That's why the team submitted a proposal for one of the National AI Resources Program's first awards. And Ganapathysubramanian said the team will make more recommendations for the program as computing requirements continue to increase.
/Open to the public. This material from the original organization/author may be of a contemporary nature and has been edited for clarity, style, and length. Mirage.News does not take any institutional position or position, and all views, positions, and conclusions expressed herein are solely those of the authors. Read the full text here.
