Random directions stochastic approximation with deterministic\n perturbations Article Swipe
YOU?
·
· 2018
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.1808.02871
We introduce deterministic perturbation schemes for the recently proposed\nrandom directions stochastic approximation (RDSA) [17], and propose new\nfirst-order and second-order algorithms. In the latter case, these are the\nfirst second-order algorithms to incorporate deterministic perturbations. We\nshow that the gradient and/or Hessian estimates in the resulting algorithms\nwith deterministic perturbations are asymptotically unbiased, so that the\nalgorithms are provably convergent. Furthermore, we derive convergence rates to\nestablish the superiority of the first-order and second-order algorithms, for\nthe special case of a convex and quadratic optimization problem, respectively.\nNumerical experiments are used to validate the theoretical results.\n
Related Topics To Compare & Contrast
- Type
- preprint
- Landing Page
- http://arxiv.org/abs/1808.02871
- https://arxiv.org/pdf/1808.02871
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4289703818