Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.04424 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911138533343232 |
|---|---|
| author | Lauand, Caio Kalil Meyn, Sean |
| author_facet | Lauand, Caio Kalil Meyn, Sean |
| contents | Simultaneous perturbation stochastic approximation (SPSA) is an approach to gradient-free optimization introduced by Spall as a simplification of the approach of Kiefer and Wolfowitz. In many cases the most attractive option is the single-sample version known as 1SPSA, which is the focus of the present paper, containing two major contributions: a modification of the algorithm designed to ensure convergence from arbitrary initial condition, and a new approach to exploration to dramatically accelerate the rate of convergence. Examples are provided to illustrate the theory, and to demonstrate that estimates from unmodified 1SPSA may diverge even for a quadratic objective function. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_04424 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Global Convergence and Acceleration for Single Observation Gradient Free Optimization Lauand, Caio Kalil Meyn, Sean Optimization and Control 62L20 Simultaneous perturbation stochastic approximation (SPSA) is an approach to gradient-free optimization introduced by Spall as a simplification of the approach of Kiefer and Wolfowitz. In many cases the most attractive option is the single-sample version known as 1SPSA, which is the focus of the present paper, containing two major contributions: a modification of the algorithm designed to ensure convergence from arbitrary initial condition, and a new approach to exploration to dramatically accelerate the rate of convergence. Examples are provided to illustrate the theory, and to demonstrate that estimates from unmodified 1SPSA may diverge even for a quadratic objective function. |
| title | Global Convergence and Acceleration for Single Observation Gradient Free Optimization |
| topic | Optimization and Control 62L20 |
| url | https://arxiv.org/abs/2509.04424 |