Spring 2021 HiPerGator Symposium highlights artificial intelligence research 

Winners of UF Research's Artificial Intelligence (AI) Research Catalyst Fund presented how they are pursuing multidisciplinary applications of artificial intelligence across the university at the Spring 2021 HiPerGator Symposium Tuesday.

Presenters covered a wide range of topics, sharing how they are using AI to tackle issues like uncovering decades of ecological change to detecting online students who are at risk of dropping out. In the afternoon, attendees and presenters interacted in Q-and-A panels. Roughly 275 people attended the event. 

“We established the AI Research Catalyst Fund as a way to encourage multidisciplinary teams of faculty and students to rapidly pursue imaginative applications of AI across the institution,” said David Norton, UF’s vice president for research. “The research shared in this Symposium clearly indicated that is what’s happening. We anticipate that this initial research will lead to significant external research funding in the future.” 

The Symposium’s focus on AI is part of a sweeping initiative to establish UF as a national leader in the field, which is widely expected to fuel future advances in research and workforce development. The projects presented at the Symposium will leverage the capabilities of HiPerGator AI, the most powerful AI supercomputer in higher education, which UF recently made widely available for teaching and research purposes.  The supercomputer, as well as the broader initiative, is made possible by a $100 million public-private partnership with Silicon Valley-based technology company NVIDIA and UF alumnus and NVIDIA co-founder Chris Malachowsky.

The university’s AI initiative empowers faculty to explore real world problems, like how to eliminate bias and create culturally inclusive communications via machine learning. Sylvia Chan-Olmsted, telecommunications professor in the College of Journalism and Media Consumer Research director, posed the question of how to find ways to increase cultural resonance in sharing information.

"Fairness has been touted as one of the most important issues for responsible AI as AI-powered systems increasingly impact human minds," she said. "At the same time, access to information is essential in today's knowledge economy and fundamental to our democracy."

Obstacles that stem from cross-cultural communication means certain groups of the population might be excluded or lack access and not be able to participate fully. To address this, Chan-Olmsted and Huan Chen, College of Journalism and Communications Advertising associate professor, will use social theories to build a culturally aware machine learning system that addresses communication in a multicultural society.   

“The Spring HiPerGator Symposium was a success for UF in many ways,” said Erik Deumens, director of UFIT Research Computing. “Having more than 275 faculty and students attend this virtual event was great."

The Symposium started three years ago as a fall semester event as a way to showcase graduate and postdoctoral work. When COVID-19 struck, the symposium transitioned online, which enabled a much wider audience to attend and learn about the research happening at UF.

"Sharing ideas with the panelists and hearing how the catalyst awardees are using machine learning will spur even more ideas for using HiPerGator AI. Plus, the attendance by faculty representing 25 universities across the U.S., shows the interest of the University of Florida's AI initiative,” Deumens said. 

Emily Cardinali April 1, 2021