Xi Yu

Xi Yu is a Researcher at the Computational Science Initiative of Brookhaven National Laboratory, with expertise in Machine Learning, Artificial Intelligence, and Computational Science. Previously, they served as a Postdoctoral Research Associate (2022–2025) and Research Intern at Mitsubishi Electric Research Laboratories (2021).

• Ph.D. in Electrical Engineering, University of Florida (2022)
• M.S. in Electrical Engineering, University of Florida (2019)

Their research spans domain generalization, adversarial robustness, information theory, and coral image segmentation, with a focus on applying neural networks and entropy functional methods to diverse challenges in AI and environmental science.

Recent publications highlight work on AI training energy efficiency, CLIP-based image-text alignment, and information bottleneck theory. Notable awards include the NYU Tandon Faculty First Look Fellowship (2024) and recognition at ICDM (2021).