Corpus ID: 13861512

Large-scale randomized-coordinate descent methods with non-separable linear constraints

@inproceedings{Reddi2015LargescaleRD,
  title={Large-scale randomized-coordinate descent methods with non-separable linear constraints},
  author={S. Reddi and A. Hefny and C. Downey and Kumar Avinava Dubey and S. Sra},
  booktitle={UAI},
  year={2015}
}
  • S. Reddi, A. Hefny, +2 authors S. Sra
  • Published in UAI 2015
  • Computer Science, Mathematics
  • We develop randomized (block) coordinate descent (CD) methods for linearly constrained convex optimization. Unlike most CD methods, we do not assume the constraints to be separable, but let them be coupled linearly. To our knowledge, ours is the first CD method that allows linear coupling constraints, without making the global iteration complexity have an exponential dependence on the number of constraints. We present algorithms and analysis for four key problem scenarios: (i) smooth; (ii… CONTINUE READING
    Randomized sketch descent methods for non-separable linearly constrained optimization.
    2
    Randomized Subspace Descent
    An almost cyclic 2-coordinate descent method for singly linearly constrained problems
    2
    Accelerated Stochastic Block Coordinate Descent with Optimal Sampling
    18
    Stochastic Coordinate Minimization with Progressive Precision for Stochastic Convex Optimization

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 52 REFERENCES
    A RANDOM COORDINATE DESCENT METHOD ON LARGE-SCALE OPTIMIZATION PROBLEMS WITH LINEAR CONSTRAINTS
    25
    A random coordinate descent algorithm for optimization problems with composite objective function and linear coupled constraints
    90
    An asynchronous parallel stochastic coordinate descent algorithm
    294
    Iteration complexity analysis of block coordinate descent methods
    87
    On the Convergence of Block Coordinate Descent Type Methods
    315
    Parallel coordinate descent methods for big data optimization
    386
    Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems
    998