Julie Nutini


PhD, Computer Science
University of British Columbia

 201 2366 Main Mall
 Vancouver, BC, V6T 1Z4
 Canada

jnutini@cs.ubc.ca

I was born and raised in the small ski town of Rossland, British Columbia, located in the West Kootenays. I graduated from the University of British Columbia (Okanagan Campus) in 2010 with my BSc in General Mathematics (with honors), and in 2012 with my MSc in Mathematical Optimization (Governor General's Gold Medal recipient). My MSc thesis, supervised by Warren Hare, focused on derivative-free optimization methods for finite minimax problems. I recently finished my PhD supervised by Mark Schmidt at the University of British Columbia in Vancouver. I work on optimization and machine learning with a focus on coordinate descent methods.

Curriculum Vitae


Publications:

J. Nutini, I. Laradji, M. Schmidt and W. Hare. Let's Make Block Coordinate Descent Go Fast: Faster Greedy Rules, Message-Passing, Active-Set Complexity, and Superlinear Convergence, submitted for publication, 2017 [pdf][slides][poster][code].

J. Nutini, M. Schmidt and W. Hare. "Active-set complexity" of proximal gradient: How long does it take to find the sparsity pattern?, Optimization Letters, 2018 [pdf] [poster].

I. Laradji, J. Nutini and M. Schmidt. Graphical Newton for Huge-Block Coordinate Descent on Sparse Graphs, NIPS Optimization Workshop, 2017 [pdf] [poster].

H. Karimi, J. Nutini and M. Schmidt. Linear Convergence of Gradient and Proximal-Gradient Methods Under the Polyak-Lojasiewicz Condition, ECML-PKDD, 2016 [pdf] [slides] [poster].

J. Nutini, B. Sepehry, I. H. Laradji, M. Schmidt, H. Koepke and A. Virani. Convergence Rates for Greedy Kaczmarz Algorithms, and Faster Randomized Kaczmarz Rules Using the Orthogonality Graph, UAI, 2016 [pdf] [poster] [code].

*K. Bigdeli, W. Hare, J. Nutini and S. Tesfamariam. Optimizing Damper Connectors for Adjacent Buildings, Optimization and Engineering, 17(1):47-75, 2016 [pdf].

J. Nutini, M. Schmidt, I. H. Laradji, M. Friedlander and H. Koepke. Coordinate Descent Converges Faster with the Gauss-Southwell Rule Than Random Selection, ICML, 2015 [pdf] [slides] [poster] [video talk].

*W. Hare, J. Nutini and S. Tesfamariam. A survey of non-gradient optimization methods in structural engineering, Advances in Engineering Software, 59:19-28, 2013 [pdf].

*W. Hare and J. Nutini. A derivative-free approximate gradient sampling algorithm for finite minimax problems, Computational Optimization and Applications, 56(1):1-38, 2013 [pdf] [slides].

* authors listed in alphabetical order


UBC Machine Learning Reading Group talks:

  • Conditional Inference and Cutset Conditioning [slides].
  • Coordinate Descent and Ascent Methods [slides].
  • Principle and Independent Component Analysis [slides].
  • Feedforward Neural Nets and Backpropagation [slides].
  • Monte Carlo Methods [slides].
  • Parallel Coordinate Descent Methods [slides].

Miscellaneous Projects

  • Research Proficiency Exam (RPE) project - Putting the Curvature Back into Sparse Solvers [pdf] [slides].