Mingrui Liu, Assistant Professor, Computer Science, College of Engineering and Computing (CEC), and Jie Xu, Professor, Systems Engineering and Operations Research (SEOR), CEC, received funding from the National Science Foundation for the project: “Collaborative Research: MFAI: Conformality-Guided Learning with Evolving and Heterogeneous Data.”
The researchers are studying Conformality-Guided Learning, a framework that uses conformal prediction signals during model training rather than only after training. Conformal prediction is a statistical approach for producing prediction sets with formal coverage guarantees.
They will develop theory and algorithms for three interconnected thrusts. First, they will analyze heterogeneous offline data and develop constrained bilevel learning methods that leverage source data and limited target-domain data to obtain valid, compact prediction sets. Second, they will study online data and develop adaptive algorithms that jointly update predictors and calibration parameters under distribution shift, including methods for subgroup and time-localized reliability. Third, they will develop communication-efficient distributed methods for selected heterogeneous and drifting environments, with a focus on common drift and performative drift.
Liu and Xu received $550,000 from NSF for this project. Funding began in Aug. 2026 and will end in late July 2029.
###
George Mason University is Virginia’s largest public research university. Located near Washington, D.C., George Mason enrolls more than 40,000 students from 130 countries and all 50 states. George Mason has grown rapidly over the past half century and is recognized for its innovation and entrepreneurship, remarkable diversity, and commitment to accessibility. In 2023, the university launched Mason Now: Power the Possible , a $1 billion comprehensive campaign to support student success, research, innovation, community, and stewardship. Learn more at GMU.EDU .