English

Efficient Implementation of Second-Order Stochastic Approximation Algorithms in High-Dimensional Problems

Optimization and Control 2019-08-26 v2 Machine Learning

Abstract

Stochastic approximation (SA) algorithms have been widely applied in minimization problems when the loss functions and/or the gradient information are only accessible through noisy evaluations. Stochastic gradient (SG) descent---a first-order algorithm and a workhorse of much machine learning---is perhaps the most famous form of SA. Among all SA algorithms, the second-order simultaneous perturbation stochastic approximation (2SPSA) and the second-order stochastic gradient (2SG) are particularly efficient in handling high-dimensional problems, covering both gradient-free and gradient-based scenarios. However, due to the necessary matrix operations, the per-iteration floating-point-operations (FLOPs) cost of the standard 2SPSA/2SG is O(p3)O(p^3), where pp is the dimension of the underlying parameter. Note that the O(p3)O(p^3) FLOPs cost is distinct from the classical SPSA-based per-iteration O(1)O(1) cost in terms of the number of noisy function evaluations. In this work, we propose a technique to efficiently implement the 2SPSA/2SG algorithms via the symmetric indefinite matrix factorization and show that the FLOPs cost is reduced from O(p3)O(p^3) to O(p2)O(p^2). The formal almost sure convergence and rate of convergence for the newly proposed approach are directly inherited from the standard 2SPSA/2SG. The improvement in efficiency and numerical stability is demonstrated in two numerical studies.

Keywords

Cite

@article{arxiv.1906.09533,
  title  = {Efficient Implementation of Second-Order Stochastic Approximation Algorithms in High-Dimensional Problems},
  author = {Jingyi Zhu and Long Wang and James C. Spall},
  journal= {arXiv preprint arXiv:1906.09533},
  year   = {2019}
}
R2 v1 2026-06-23T10:00:56.505Z