Luke Olson

Luke Olson is a Professor in the Department of Computer Science at the University of Illinois at Urbana-Champaign (UIUC), part of the College of Engineering. He holds an affiliate appointment in the Department of Mechanical Science and Engineering. His research focuses on numerical methods, high-performance computing, and parallel algorithms, particularly algebraic multigrid (AMG) solvers and sparse matrix computations. He leads the development of open-source libraries like PyAMG and RAPtor, advancing computational tools for scientific and engineering applications.

Education: Ph.D. in Applied Mathematics from the University of Colorado Boulder (2003), M.S. in Mathematics from the University of Iowa (1999), and B.A. in Mathematics and Physics from Luther College (1997).

Research Interests:

• Algebraic multigrid methods and preconditioners
• High-performance computing and parallel algorithms
• Numerical solutions to partial differential equations
• Scientific computing and software development
• Machine learning integration in numerical simulations

Recent Articles Trends: Recent work merges machine learning with traditional numerical methods, such as neural network closures for turbulent combustion and reduced basis approximations using neural networks. Ongoing contributions include optimizing multigrid methods for exascale architectures and enhancing communication efficiency in distributed systems.

Awards: Recognized with the NSF CAREER Award (2007), UIUC Campus Award for Excellence in Teaching (2024), and the Donald Biggar Willett Faculty Scholar distinction (2016). Active in conference organization, including the Copper Mountain Conference on Multigrid Methods.

Teaching & Grants: Teaches courses like Numerical Methods for PDEs and Scientific Machine Learning. Leads projects funded by NSF and industry collaborations, emphasizing education innovation through the AE3 fellowship (2014–2016).

Labs/Teams: Directs the Scientific Computing Group at UIUC and contributes to interdisciplinary initiatives like the Center for Exascale-enabled Scramjet Design (CEESD).