image: The process of the adaptive immune system developing solutions to remove antigens inspires the computational system model for developing solutions to remove adversaries. Information collected from the power grid and physical constraints are used in the practical implementation of the computational system. view more
Credit: Illinois Institute of Technology
CHICAGO—July 25, 2023—For his groundbreaking research in fortifying artificial intelligence systems with insights gained from the human immune system, Ren Wang of Illinois Institute of Technology has received the prestigious Ralph E. Powe Junior Faculty Enhancement Award from Oak Ridge Associated Universities (ORAU). Wang’s research may be used in the future to strengthen AI systems, making them more robust and resilient.
As AI has increasingly permeated our daily lives through technologies, such as ChatGPT’s natural language processing and facial recognition in smartphones, the technology’s vulnerability to attacks and errors has been exposed—including everything from self-driving cars crashing to T-shirts designed to foil facial recognition.
“This is a very dangerous phenomenon,” says Wang, an assistant professor in the Department of Electrical and Computer Engineering. “A very small perturbation can lead to a totally wrong answer. For example, the AI model might predict a speed limit sign when there is actually a stop sign.”
Because AI systems are like a black box—that is, it’s often not possible to know why an AI system makes any given decision—it’s difficult to develop techniques to keep them safe. To address this, Wang borrowed insights from the immune system—specifically the adaptive mechanisms of B cells, which are integral components of our immune defenses.
“B cells generate antibodies to defend against certain types of attack, and that attack could be any antigen,” says Wang. “We hope to learn from this whole process and to capture some important patterns that we can use to improve the AI system.”
Wang’s research is not limited to broad-based AI applications; he’s particularly focused on AI-driven power system applications, such as power system control and stability analysis, requiring a high degree of robustness. By integrating physical constraints into his immune-inspired learning approach, Wang aims to create robust AI models for complex power grids that necessitate less data and have a refined search space.
“I believe that this is a very novel idea,” says Wang, “and I believe this adaptive immune-inspired system will be very powerful.”
Wang is among 35 recipients selected from 167 applicants in 2023 for the Powe Award, which provides seed money for junior faculty members demonstrating exceptional promise in scientific research.
“It is a tremendous honor to have been selected as one of this year’s awardees, and I am immensely grateful,” says Wang.
The award follows another notable achievement by Wang, who recently received a Computer and Information Science and Engineering Research Initiation Initiative award from the National Science Foundation.