Corpus ID: 16264178

Doubly Stochastic Primal-Dual Coordinate Method for Empirical Risk Minimization and Bilinear Saddle-Point Problem

@article{Yu2015DoublySP,
  title={Doubly Stochastic Primal-Dual Coordinate Method for Empirical Risk Minimization and Bilinear Saddle-Point Problem},
  author={Adams Wei Yu and Qihang Lin and Tianbao Yang},
  journal={arXiv: Learning},
  year={2015}
}
We proposed a doubly stochastic primal-dual coordinate (DSPDC) optimization algorithm for empirical risk minimization, which can be formulated as a bilinear saddle-point problem. In each iteration, our method randomly samples a block of coordinates of the primal and dual solutions to update. The convergence of our method is established in both the distance from the current iterate to the optimal solution and the primal-dual objective gap. We show that the proposed method has a lower overall… Expand

Figures and Tables from this paper

Dual Iterative Hard Thresholding
Stochastic Primal-Dual Method for Empirical Risk Minimization with $\mathcal{O}(1)$ Per-Iteration Complexity.
Level-Set Methods for Finite-Sum Constrained Convex Optimization
Block-proximal methods with spatially adapted acceleration
...
1
2
...

References

SHOWING 1-10 OF 59 REFERENCES
Excessive Gap Technique in Nonsmooth Convex Minimization
  • Y. Nesterov
  • Mathematics, Computer Science
  • SIAM J. Optim.
  • 2005
A Proximal Stochastic Gradient Method with Progressive Variance Reduction
A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging
  • A. Chambolle, T. Pock
  • Computer Science, Mathematics
  • Journal of Mathematical Imaging and Vision
  • 2010
...
1
2
3
4
5
...