Universality in Numerical Computation with Random Data: Case Studies, Analytical Results and Some Speculations

  • Percy DeiftEmail author
  • Thomas Trogdon
Conference paper
Part of the Abel Symposia book series (ABEL, volume 13)


We discuss various universality aspects of numerical computations using standard algorithms. These aspects include empirical observations and rigorous results. We also make various speculations about computation in a broader sense.



One of the authors (P.D.) would like to thank the organizers for the invitation to participate in the Abel Symposium 2016 “Computation and Combinatorics in Dynamics, Stochastics and Control”. During the symposium he gave a talk on a condensed version of the paper below.


  1. 1.
    Bakhtin, Y., Correll, J.: A neural computation model for decision-making times. J. Math. Psychol. 56, 333–340 (2012)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Deift, P., Li, L.C., Nanda, T., Tomei, C.: The Toda flow on a generic orbit is integrable. Commun. Pure Appl. Math. 39(2), 183–232 (1986)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Deift, P., Menon, G., Olver, S., Trogdon, T.: Universality in numerical computations with random data. Proc. Natl. Acad. Sci. USA 111(42):14973–14978 (2014)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Deift, P., Nanda, T., Tomei, C.: Ordinary differential equations and the symmetric eigenvalue problem. SIAM J. Num. Anal. 20, 1–22 (1983)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Deift, P., Trogdon, T.: Universality for the Toda algorithm to compute the largest eigenvalue of a random matrix. Commun. Pure Appl. Math. arXiv:1604.07384Google Scholar
  6. 6.
    Deift, P., Trogdon, T.: Universality for eigenvalue algorithms on sample covariance matrices. arXiv:1701.01896Google Scholar
  7. 7.
    Markus, G.: Face it, your brain is a computer. New York Times, 27 June 2015Google Scholar
  8. 8.
    Mehta, M.L.: Random matrices, 3rd edn. Elsevier, Amsterdam, 2004zbMATHGoogle Scholar
  9. 9.
    Moser, J.: Three integrable Hamiltonian systems connected with isospectral deformations. Adv. Math. 16(2), 197–220 (1975)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Perret, A., Schehr. G.: Near-extreme eigenvalues and the first gap of Hermitian random matrices. J. Stat. Phys. 156(5), 843–876 (2014)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Pfrang, C.W., Deift, P., Menon, G.: How long does it take to compute the eigenvalues of a random symmetric matrix? Random Matrix Theory Interact. Part. Syst. Integr. Syst. MSRI Publ. 65, 411–442 (2014)MathSciNetzbMATHGoogle Scholar
  12. 12.
    Sagun, L., Trogdon, T., LeCun, Y.: Universal halting times in optimization and machine learning. arXiv:1511.06444Google Scholar
  13. 13.
    Symes, W.W.: The QR algorithm and scattering for the finite nonperiodic Toda lattice. Phys. D Nonlinear Phenom. 4(2), 275–280 (1982)MathSciNetCrossRefGoogle Scholar
  14. 14.
    University of Colorado at Boulder: Human brain region functions like a digital computer. Science Daily, 6 Oct 2006Google Scholar
  15. 15.
    Witte, N.S., Bornemann, F., Forrester, P.J.: Joint distribution of the first and second eigenvalues at the soft edge of unitary ensembles. Nonlinearity 26(6), 1799–1822 (2013)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Courant InstituteNew York UniversityNew YorkUSA
  2. 2.University of California, IrvineIrvineUSA

Personalised recommendations