معرفی
X.Y. Han is an Assistant Professor in the Department of Operations Management & Applied AI at the Booth School of Business, University of Chicago. Their research bridges artificial intelligence, optimization, and human-centered systems, aiming to enhance professional productivity through scientific understanding of AI behaviors.
Education:
- BSE in Operations Research and Financial Engineering, Princeton University
- PhD in Operations Research and Information Engineering, Cornell University
- Post-doctoral & MS in Statistics, Stanford University
Dr. Han's research focuses on demystifying black-box AI systems by uncovering fundamental phenomena such as neural collapse through large-scale experiments and geometric analysis. They also develop novel optimization algorithms grounded in nonsmooth, constrained, and nonconvex theory, integrating modern computational practices. Applications target real-world impact through human-in-the-loop tools developed in collaboration with institutions like the Frick Art Reference Library, Veolia North America LLC, and the Surgical Data Science Collective.
Their published work, particularly on neural collapse, has appeared in top venues including PNAS and ICLR, revealing deep structural patterns in neural network training dynamics. These studies span disciplines such as machine learning, optimization, and statistical learning, with subfields including feature geometry, generalization, and training convergence.
Scientific Awards:
- ICLR 2022 Outstanding Paper Award
- ICCOPT 2022 Best Paper Prize for Young Researchers (Finalist)
Dr. Han has been advised by leading scholars including Adrian S. Lewis and David L. Donoho. While no specific grants are listed, their collaborative projects suggest externally supported research efforts. Their work continues to advance the scientific foundations of AI and optimization, with emphasis on interpretability and practical utility.
Dr. Han is involved in interdisciplinary teams focused on surgical data science and industrial AI applications, contributing technical expertise in algorithm design and data analysis.
