English

Improved Analysis and Rates for Variance Reduction under Without-replacement Sampling Orders

Machine Learning 2021-10-28 v2 Optimization and Control

Abstract

When applying a stochastic algorithm, one must choose an order to draw samples. The practical choices are without-replacement sampling orders, which are empirically faster and more cache-friendly than uniform-iid-sampling but often have inferior theoretical guarantees. Without-replacement sampling is well understood only for SGD without variance reduction. In this paper, we will improve the convergence analysis and rates of variance reduction under without-replacement sampling orders for composite finite-sum minimization. Our results are in two-folds. First, we develop a damped variant of Finito called Prox-DFinito and establish its convergence rates with random reshuffling, cyclic sampling, and shuffling-once, under both convex and strongly convex scenarios. These rates match full-batch gradient descent and are state-of-the-art compared to the existing results for without-replacement sampling with variance-reduction. Second, our analysis can gauge how the cyclic order will influence the rate of cyclic sampling and, thus, allows us to derive the optimal fixed ordering. In the highly data-heterogeneous scenario, Prox-DFinito with optimal cyclic sampling can attain a sample-size-independent convergence rate, which, to our knowledge, is the first result that can match with uniform-iid-sampling with variance reduction. We also propose a practical method to discover the optimal cyclic ordering numerically.

Keywords

Cite

@article{arxiv.2104.12112,
  title  = {Improved Analysis and Rates for Variance Reduction under Without-replacement Sampling Orders},
  author = {Xinmeng Huang and Kun Yuan and Xianghui Mao and Wotao Yin},
  journal= {arXiv preprint arXiv:2104.12112},
  year   = {2021}
}

Comments

Accepted by NeurIPS 2021

R2 v1 2026-06-24T01:29:35.190Z