Google Scholar


Fred (Farbod) Roosta
School of Mathematics and Physics
University of Queensland, Brisbane, Australia
International Computer Science Institute, Berkeley, USA

Email: fred.roostauq.edu.au


Research Interests

  • Machine Learning
  • Numerical Optimization
  • Randomized Algorithms
  • Computational Statistics
  • Numerical Analysis and Scientific Computing


  • January, 2019: Straight from Down Under, we bring you DINGO, our new communication-efficient distributed Newton-type algorithm for invex problems.
  • December, 2018: Our paper on acceleration of sub-sampled Newton methods using GPUs has been accepted at the SIAM International Conference on Data Mining - SDM19 (22.7% acceptance rate).
  • December, 2018: I will be serving as a senior program committee member for the 28th International Joint Conference on Artificial Intelligence (IJCAI 2019).
  • October, 2018: After 19 months, 7 reviewers, and several rounds of reviews, our paper on "Sub-sampled Newton methods" finally appeared in Mathematical Programming.
  • October, 2018: In this paper, which is an extension of the results of our prior work, we study the invariance kernel and the weight distributions during the training of MLPs.
  • October, 2018: Our paper on Newton-MR is now available. Newton-MR is a second-order method that looks very similar to the classical Newton's method, but extends its application range far beyound smoothness and convexity.
  • September, 2018: The video of my talk on "Randomized Analysis and Deep Learning" as part of the Randomized Numerical Linear Algebra and Applications workshop at the Simons Institute for the Theory of Computing is available here.
  • September, 2018: Our paper on GIANT, a distributed optimization algorithm for convex problems, has been accepted in NIPS 2018.
  • August, 2018: The videos of my two talks on the introduction to (stochastic) second-order optimization methods as part of the Foundations of Data Science Boot Camp at the Simons Institute for the Theory of Computing are available: Part I (Convex) and Part II (Non-Convex).
  • August, 2018: Pytorch code with MPI support for Newton-ADMM is now available here.
  • June, 2018: I was selected for the ICML 2018 Top 10 Reviewer award! It is like a "fair-play award" in some sports, or even a "good-behaviour award" in elementary school, but nonetheless, I am going to brag about it :-)
  • May, 2018: We are represented at ICML 2018 with 2 papers! (this, which is about inference on random graphs and this, which studies some statistical properties of artificial neural nets).

Preprints and Publications




  • Optimization Methods for Inverse Problems
    Nan Ye, Fred Roosta, and Tiangang Cui
    2017 MATRIX Annals, Editors: David R. Wood, Jan de Gier, Cheryl E. Praeger, Terence Tao. MATRIX Book Series, Volume 2, Springer (to appear)


  • Parallel Local Graph Clustering
    Julian Shun, Fred Roosta, Kimon Fountoulakis and Michael W. Mahoney
    Proc. of the VLDB Endowment, 9(12), pp. 1041-1052, 2016.




  • Yang Liu (PhD, 2017 - present )
  • Robert Salomone (PhD, Completed 2019, Now Postdoc at UNSW)


I always look for enthusiastic PhD students...

So, if (a) and (b), then feel free to drop me a line with your current CV and a few words regarding your research/background.

(a) Your research interests lie anywhere in the union of Machine Learning, Numerical Optimization, Randomized Algorithms, Computational Statistics, Scientific Computing, or Numerical Linear Algebra.

You are looking to do a PhD degree in one of the world's leading and highly ranked research universities in beautiful Australia.

Selected Presentations


  • INFORMS Annual Meeting

    Phoenix, Arizona, November, 2018



  • Randomized Numerical Linear Algebra and Applications - Simons Institute for the Theory of Computing

    Berkeley, California, September, 2018

  • Foundations of Data Science Boot Camp- Simons Institute for the Theory of Computing

    Berkeley, California, August, 2018

  • International Symposium on Mathematical Programming (ISMP)

    Bordeaux, France, July, 2018

  • International Conference on Econometrics and Statistics (EcoSta)

    Hong Kong, June, 2018

  • International Conference on Artificial Intelligence and Statistics (AISTATS)

    Lanzarote, Canary Islands, April, 2018

  • International Conference on High Performance Scientific Computing (HPSC)

    Hanoi, Vietnam, March, 2018

  • SIAM Annual Meeting

    Pittsburgh, Pennsylvania, July, 2017

  • Matrix Program on Computational Inverse Problems

    Creswick, Australia, June, 2017

  • SIAM Conference on Optimization

    Vancouver, Canada, May, 2017

  • SIAM Conference on Computational Science and Engineering

    Alanta, Georgia,Atlanta, Georgia, February, 2017

  • Computational and Methodological Statistics (CMStatistics)

    Seville, Spain, December, 2016

  • Pacific Institute for the Mathematical Sciences (SCAIM Seminar)

    Vancouver, British Columbia, September, 2016

  • Institute of Applied Mathematics (IAM)

    Vancouver, British Columbia, September, 2016

  • Virginia Tech Math Colloquium

    Blacksburg, Virginia, September, 2016

  • SAMSI Program on Optimization (Invited Plenary Talk)

    Research Triangle Park, North Carolina, August, 2016

  • Recent Advances in Randomized Numerical Linear Algebra

    National Institute of Informatics, Shonan Village Center, Japan, July, 2016

  • Workshop on Algorithms for Modern Massive Data Sets (MMDS)

    Berkeley, California, June, 2016

  • SIAM Conference on Uncertainty Quantification

    EPFL, Lausanne, Switzerland, April, 2016

  • Matrix Computations and Scientific Computing Seminar

    Berkeley, California, March, 2016

  • AMPLab (All Hands Meeting)

    Berkeley, California, September, 2015

  • Sparse Representations, Numerical Linear Algebra, and Optimization

    BIRS, Alberta, October, 2014

  • SIAM Conference on Optimization

    San Diego, California, May, 2014

  • Pacific Institute for the Mathematical Sciences (SCAIM Seminar)

    Vancouver, British Columbia, November, 2013