Imagine a world where a farmer’s smart phone predicts the perfect day to harvest. Or a governor can dial up exactly how to enhance food security prior to a hurricane.
It would take seamless access to a highly technical artificial intelligence (AI) infrastructure, but Ohio State’s Dhabaleswar K. (DK) Panda is working to get us there.
“If you look at AI, it’s become very important but it is limited to only advanced technical people,” said Panda, professor of computer science and engineering at Ohio State. “How do we take it to the masses? We want to create a plug-and-play AI that will be democratized so anybody can use it.”
Panda is leading a National Science Foundation institute to do exactly that. The institute is called ICICLE (The AI Institute for Intelligent Cyberinfrastructure with Computational Learning in the Environment) and features a team of 46 academic researchers and staff scientists from 13 organizations, including many from Ohio State, all working to build a cyberinfrastructure to make AI accessible to society.
While this project is in the early stages, Panda has a track record of developing computational tools for communities at large. He is a rock star (no, seriously) in the high performance computing world, having developed supercomputing software libraries that are being used by more than 3,000 organizations in 89 countries.
“Traditionally our work has been targeted to purely scientific computing,” he said. “But now with AI coming, it requires similar kinds of answers. My research team and I have been exploring this new direction — how to combine HPC with AI with big data — and that has led to a lot of foundational developments.”
“DK Panda represents a unique powerhouse at Ohio State,” Berger-Wolf said. “He is the world’s leader in high performance computing (HPC) and particularly developing this intelligent adaptive HPC. I can think of few people, if any, in the world who could have put together a team like (ICICLE).
“Ohio State is the one place you can do a project like this. We have the right people across the spectrum of necessary expertise, and they’re willing and able to answer big questions.”
ICICLE will initially build and prove its AI cyberinfrastructure in three domains: smart foodsheds, digital agriculture and animal ecology. Long-term, this same cyberinfrastructure could be adapted to many arenas. In health care, for instance, it could mean faster medicinal development or quicker diagnosis of diseases, such as cancer.
In the process of building this AI infrastructure, Panda said there will be a strong focus on workforce development, training the next generation for the jobs of tomorrow.
“This is very important,” Panda said. “If you think of this vision, to allow people anywhere to use AI, there’s a lot of work needed to build the software and design these systems. Through this process, we’ll be training the next generation’s workforce for a new generation of jobs.”