David Persson

Current Positions

• Courant Instructor/Assistant Professor, NYU Courant Institute of Mathematical Sciences
• Flatiron Research Fellow, Flatiron Institute, Center for Computational Mathematics

About me

I am a Courant Instructor at the Courant Institute and a Research Fellow at the Flatiron Institute. My research focuses on the design and analysis of numerical linear algebra algorithms, with particular interests in randomized methods, hierarchical matrices, matrix functions, low-rank approximation, trace estimation, and applications to machine learning. I received my PhD in mathematics from EPFL, where I was advised by Daniel Kressner. My work on low-rank approximations of matrix functions was recognized with the second Leslie Fox Prize. I have also worked on computing the polar factor in the Muon algorithm, which received an Outstanding Paper Honorable Mention at ICLR 2026.

Curriculum Vitae

Download my CV (PDF)

PhD Thesis

Title: Randomized low-rank approximation and its applications
Institution: EPFL, 2024
📄 Download PDF
📘 Thesis landing page on EPFL InfoScience

Publications

• N. Amsel, P. Avi, F.D. Keles, C. Hegde, C. Musco, C. Musco, and D. Persson, Query Efficient Structured Matrix Learning, Conference on Learning Theory (to appear), (2026).
• N. Amsel, D. Persson, C. Musco, and R.M. Gower, The Polar Express: Optimal matrix-sign methods and their application within the Muon method, International Conference on Learning Representations (to appear), (2026).
• N. Amsel, T. Chen, F.D. Keles, D. Halikias, C. Musco, C. Musco, and D. Persson, Quasi-optimal hierarchically semi-separable matrix approximation, SIAM J. Matrix Anal. Appl., 47 (2026), pp. 586–621.
• D. Persson, T. Chen, and C. Musco, Randomized block-Krylov subspace methods for low-rank approximation of matrix functions, Linear Algebra Appl., vol. 741, pp. 32–65, 2026.
• D. Kressner, D. Persson, and A. Uschmajew, On the randomized SVD in infinite dimensions, Linear Algebra Appl., vol. 735, pp. 259–286, 2026.
• D. Persson, N. Boullé, and D. Kressner, Randomized Nyström approximation of non-negative self-adjoint operators, SIAM J. Data Sci. 7 (2025), pp. 670-698.
• T. Chen, F.D. Keles, D. Halikias, C. Musco, C. Musco, and D. Persson, Near-optimal hierarchical matrix approximation from matrix-vector products, in Proceedings of the 2025 Annual ACM-SIAM Symposium on Discrete Algorithms (SODA), 2025, pp. 2656–2692.
• D. Persson, R.A. Meyer, and C. Musco, Algorithm-agnostic low-rank approximation of operator monotone matrix functions, SIAM J. Matrix Anal. Appl., 46 (2025), pp. 1–21.
• D. Persson and D. Kressner, Randomized low-rank approximation of monotone matrix functions, SIAM J. Matrix Anal. Appl., 44 (2023), pp. 894–918.
• D. Persson, A. Cortinovis, and D. Kressner, Improved variants of the Hutch++ algorithm for trace estimation, SIAM J. Matrix Anal. Appl., 43 (2022), pp. 1162–1185.

Contact

Email: dup210 [at] nyu.edu or dpersson [at] flatironinstitute.org

Find Me Online

GitHub | Google Scholar | LinkedIn