Ying Cui Receives NSF CAREER Award for Optimization Research
Ying Cui, an assistant professor in the Department of Industrial Engineering and Operations Research at University of California, Berkeley, has received the Faculty Early Career Development (CAREER) Award from the National Science Foundation.
The CAREER Award is among the NSF’s most prestigious honors for early-career faculty. It will provide approximately $631,492 over five years to support Cui’s project, “Foundations and Algorithms for Nonconvex and Nonsmooth Optimization: From Local Solvers to Global Certification.” Her research focuses on advancing mathematical tools that underpin modern artificial intelligence and decision-making systems.
Advancing the Computational Foundations of Modern Decision-Making Systems
Optimization serves as a central mathematical framework for artificial intelligence, enabling systems to learn from data and make decisions. However, real-world applications often present challenges that classical optimization methods are not designed to address. These include nonconvexity, in which complex solution landscapes can trap algorithms in suboptimal points; nonsmoothness, where structural irregularities complicate analysis and reduce algorithmic stability; and stochasticity, or noise in data, which can disrupt the convergence of learning processes.
Cui’s research aims to address these challenges by developing rigorous theoretical foundations alongside scalable algorithms tailored to such environments. Her work will focus on designing novel methods capable of navigating ill-conditioned problems while achieving performance comparable to that of more structured and well-behaved systems.
Algorithmic Innovation and Practical Impact
The project introduces several advances in optimization and machine learning. Cui will develop scalable first-order and second-order methods that leverage underlying problem structure and improve computational efficiency. In addition, her work explores techniques for certificating global optimality through a homotopic sketching framework, providing reliable guarantees that models reach high-quality solutions. The project also includes the development of open-source computational tools intended to make these methods accessible to a broader research community.
Integrating Research and Education
In parallel with its research contributions, the CAREER award supports an integrated educational initiative. Cui plans to translate her findings into instructional materials for a wide range of learners, including K-12 students, community college educators and graduate-level researchers.
The project, which is expected to run through March 2031, aims to expand access to advanced optimization concepts while preparing the next generation of scholars to address emerging challenges in artificial intelligence and engineering.