Ethan N. EpperlyView profile
Research Fellow
Ethan N. Epperly is a Miller Research Fellow in the Department of Mathematics at the University of California, Berkeley, where he conducts cutting-edge research in applied mathematics with a focus on computational techniques for large-scale problems. Dr. Epperly received his PhD in Applied and Computational Mathematics from Caltech, where his research was supported by a Department of Energy Computational Science Graduate Fellowship. His educational background established a strong foundation in both theoretical and applied mathematics. His primary research interests include randomized and quantum algorithms, scientific computing, and large-scale machine learning. Dr. Epperly specializes in designing computational techniques for solving large-scale problems in machine learning, quantum information, and scientific computing, with particular expertise in kernel matrix approximation, low-rank approximation, and numerical linear algebra problems. His work bridges theoretical analysis with practical computational efficiency. Epperly's recent publications demonstrate a strong focus on developing efficient randomized algorithms for matrix computations. His research shows how randomized approaches can achieve accuracy and stability comparable to classical methods while offering significant computational advantages, particularly in settings where computational resources are limited. His work on Krylov subspace methods, Cholesky decomposition variants, and trace estimation has advanced the field of numerical linear algebra. Hertz foundation fellowship finalist Thomas A. Tisch Prize for Graduate Teaching in CMS W. P. Carey & Co. Prize in Applied Mathematics SIAM Student Paper Prize Department of Energy Computational Science Graduate Fellowship As a Miller Research Fellow, Dr. Epperly collaborates with leading researchers including Joel A. Tropp, Robert J. Webber, and Yifan Chen. His work has significant implications for machine learning applications requiring efficient handling of large-scale matrix computations, with potential applications across scientific computing and quantum information processing.







