Drexel University will receive $1.4 million from the US Department of War for its role in a five-year, $7.5 million Multidisciplinary University Research Initiative (MURI) project.
Yuan An, PhD
The project, named "SOLIR: Stochastic Ontology for Learning: Information Theory Meets Rule-Based Reasoning", combines information theory, distributed learning and inference, and ontology engineering to enable modern AI systems to make better decisions in complex and changing environments, according to Yuan An, PhD, co-PI on SOLIR and associate professor in the information science department of Drexel’s Nick Howley College of Engineering and Computing.
“The SOLIR project aims to integrate the successes of modern AI and machine learning systems such as LLMs with the reasoning guarantees of formal reasoning engines such as ontologies,” said An. “One of the significant impacts is that the reasoning process of AI will be explicit and the results will be trustworthy.”
An will lead the Drexel research portion: the development of theoretical frameworks and practical implementations for stochastic ontology, or representing knowledge in raw data sources in structured formats that allow for the capture of uncertainty and adaptation to evolving knowledge.
Over the course of the project, An will work with the team as it develops novel solutions and trains a new generation of AI researchers.
“I will collaborate with the team members to measure the uncertainties in inference, representation and generation by applying methods in information theory, distributed computing and ontology engineering,” said An. “We will test these methods in real-world applications in mission-critical domains.”
SOLIR, one of only 20 2026 MURI submissions awarded from an initial pool of 451, is led by principal investigator Salim El Rouayheb, an associate professor of electrical and computer engineering at Rutgers University, where the project team held their initial workshop in October.
Learn More About SOLIR