{"ID":2874733,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.04424","arxiv_id":"2509.04424","title":"Global Convergence and Acceleration for Single Observation Gradient Free Optimization","abstract":"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.","short_abstract":"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, containin...","url_abs":"https://arxiv.org/abs/2509.04424","url_pdf":"https://arxiv.org/pdf/2509.04424v1","authors":"[\"Caio Kalil Lauand\",\"Sean Meyn\"]","published":"2025-09-04T17:45:54Z","proceeding":"math.OC","tasks":"[\"math.OC\"]","methods":"[\"LoRA\"]","has_code":false}