Corpus ID: 8890061

Coordinate Descent Converges Faster with the Gauss-Southwell Rule Than Random Selection

@inproceedings{Nutini2015CoordinateDC,
  title={Coordinate Descent Converges Faster with the Gauss-Southwell Rule Than Random Selection},
  author={Julie Nutini and Mark W. Schmidt and Issam H. Laradji and Michael P. Friedlander and Hoyt A. Koepke},
  booktitle={ICML},
  year={2015}
}
  • Julie Nutini, Mark W. Schmidt, +2 authors Hoyt A. Koepke
  • Published in ICML 2015
  • Computer Science, Mathematics
  • There has been significant recent work on the theory and application of randomized coordinate descent algorithms, beginning with the work of Nesterov [SIAM J. Optim., 22(2), 2012], who showed that a random-coordinate selection rule achieves the same convergence rate as the Gauss-Southwell selection rule. This result suggests that we should never use the Gauss-Southwell rule, as it is typically much more expensive than random selection. However, the empirical behaviours of these algorithms… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 126 CITATIONS

    Efficient Greedy Coordinate Descent for Composite Problems

    VIEW 15 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    Convergence Rates for Greedy Kaczmarz Algorithms, and Faster Randomized Kaczmarz Rules Using the Orthogonality Graph

    VIEW 6 EXCERPTS
    CITES BACKGROUND, RESULTS & METHODS

    Accelerating Greedy Coordinate Descent Methods

    VIEW 1 EXCERPT
    CITES BACKGROUND

    FILTER CITATIONS BY YEAR

    2013
    2020

    CITATION STATISTICS

    • 17 Highly Influenced Citations

    • Averaged 21 Citations per year from 2018 through 2020

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 41 REFERENCES

    Accelerated, Parallel, and Proximal Coordinate Descent

    VIEW 2 EXCERPTS

    Parallel coordinate descent methods for big data optimization

    VIEW 1 EXCERPT

    A coordinate gradient descent method for nonsmooth separable minimization

    VIEW 3 EXCERPTS