English

Stochastic Variance-Reduced Hamilton Monte Carlo Methods

Machine Learning 2020-10-20 v2 Machine Learning Computation

Abstract

We propose a fast stochastic Hamilton Monte Carlo (HMC) method, for sampling from a smooth and strongly log-concave distribution. At the core of our proposed method is a variance reduction technique inspired by the recent advance in stochastic optimization. We show that, to achieve ϵ\epsilon accuracy in 2-Wasserstein distance, our algorithm achieves O~(n+κ2d1/2/ϵ+κ4/3d1/3n2/3/ϵ2/3)\tilde O(n+\kappa^{2}d^{1/2}/\epsilon+\kappa^{4/3}d^{1/3}n^{2/3}/\epsilon^{2/3}) gradient complexity (i.e., number of component gradient evaluations), which outperforms the state-of-the-art HMC and stochastic gradient HMC methods in a wide regime. We also extend our algorithm for sampling from smooth and general log-concave distributions, and prove the corresponding gradient complexity as well. Experiments on both synthetic and real data demonstrate the superior performance of our algorithm.

Keywords

Cite

@article{arxiv.1802.04791,
  title  = {Stochastic Variance-Reduced Hamilton Monte Carlo Methods},
  author = {Difan Zou and Pan Xu and Quanquan Gu},
  journal= {arXiv preprint arXiv:1802.04791},
  year   = {2020}
}

Comments

23 pages, 3 figures, 4 tables. In ICML 2018

R2 v1 2026-06-23T00:21:23.859Z