English

Stochastic Primal-Dual Method for Empirical Risk Minimization with $\mathcal{O}(1)$ Per-Iteration Complexity

Optimization and Control 2018-11-06 v1 Machine Learning

Abstract

Regularized empirical risk minimization problem with linear predictor appears frequently in machine learning. In this paper, we propose a new stochastic primal-dual method to solve this class of problems. Different from existing methods, our proposed methods only require O(1) operations in each iteration. We also develop a variance-reduction variant of the algorithm that converges linearly. Numerical experiments suggest that our methods are faster than existing ones such as proximal SGD, SVRG and SAGA on high-dimensional problems.

Keywords

Cite

@article{arxiv.1811.01182,
  title  = {Stochastic Primal-Dual Method for Empirical Risk Minimization with $\mathcal{O}(1)$ Per-Iteration Complexity},
  author = {Conghui Tan and Tong Zhang and Shiqian Ma and Ji Liu},
  journal= {arXiv preprint arXiv:1811.01182},
  year   = {2018}
}

Comments

NIPS 2018

R2 v1 2026-06-23T05:02:59.041Z