Tim Reid

Applied Mathematics Ph.D. Student

About

I am an applied mathematics Ph.D. student graduating in Spring 2022. I have research experience in probabilistic numerics, Bayesian neural networks, uncertainty quantification, and numerical linear algebra.

The main research area for my Ph.D. is probabilistic numerics. Probabilistic numerics is the application of statistical ideas to existing numerical methods to obtain a statistical model of computational error. The statistical error model is useful because it can be propagated through a computational pipeline more easily than a traditional error bound. The probabilistic numerical methods I focus on are linear solvers.

While pursuing my Ph.D., I participated in summer internships at Sandia National Laboratories and MIT Lincoln Laboratory. At Sandia, I examined different methods of incorporating physical model parameters in a surrogate model for Earth science data. At Lincoln Laboratory, I researched Bayesian methods to improve calibration of neural networks used in classification problems.

Professional Information

Papers

Journal Articles

  1. H. Al Daas, G. Ballard, P. Cazeaux, E. Hallman, A. Miedlar, M. Pasha, T. W. Reid, and A. K. Saibaba, Randomized algorithms for rounding in the tensor-train format, (2021)
    Link to preprint
  2. T. W. Reid, I. C. F. Ipsen, J. Cockayne, and C. J. Oates, BayesCG as an uncertainty aware version of CG, (2021)
    Link to preprint
    Link to related software
  3. J. Cockayne, I. C. F. Ipsen, C. J. Oates, and T. W. Reid, Probabilistic iterative methods for linear systems, Journal of Machine Learning Research, 22 (2021), pp. 1–34,
    Link to journal
  4. D. M. Anderson, M. Corsaro, J. Horton, T. Reid, and P. Seshaiyer, Tear film dynamics with blinking and contact lens motion, Mathematical Medicine and Biology: A Journal of the IMA, 38 (2021), pp. 355--395
    Link to journal
    Please contact me if you want a PDF of this article

Technical Report

  1. T. Reid, C. Safta, A. Gorodetsky, J. Jakeman, and K. Sargsyan, Implementing physical dependence in the functional tensor train, in Computer Science Research Institute Summer Proceedings 2019, M. Powell and M. J. Parks, eds., Technical Report SAND2020-9969R, Sandia National Laboratories, 2020, pp. 55--65

Contact Information