A Parameter-Free Zeroth-Order Algorithm for Decentralized Stochastic Convex Optimization
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910054672760832 |
|---|---|
| author | Chen, Jiawei Rogozin, Alexander |
| author_facet | Chen, Jiawei Rogozin, Alexander |
| contents | We consider decentralized stochastic convex optimization on connected network, in which gradients of agents are unavailable and each agent can query only noisy function values of its own local objective. The goal is to minimize the average objective over a compact convex domain using only local two point zeroth-order oracles and peer-to-peer communication. We propose a decentralized POEM method (D-POEM) that combines symmetric two point smoothing with adaptive radius and stepsize rules, thereby avoiding prior knowledge of the Lipschitz constant and diameter. For convex Lipschitz continuous objectives, we prove an convergence rate that separates a centralized optimization term from a network disagreement term. We further conduct the numerical experiments to demonstrate POEM outperforms existing distributed zeroth-order method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_15219 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Parameter-Free Zeroth-Order Algorithm for Decentralized Stochastic Convex Optimization Chen, Jiawei Rogozin, Alexander Optimization and Control We consider decentralized stochastic convex optimization on connected network, in which gradients of agents are unavailable and each agent can query only noisy function values of its own local objective. The goal is to minimize the average objective over a compact convex domain using only local two point zeroth-order oracles and peer-to-peer communication. We propose a decentralized POEM method (D-POEM) that combines symmetric two point smoothing with adaptive radius and stepsize rules, thereby avoiding prior knowledge of the Lipschitz constant and diameter. For convex Lipschitz continuous objectives, we prove an convergence rate that separates a centralized optimization term from a network disagreement term. We further conduct the numerical experiments to demonstrate POEM outperforms existing distributed zeroth-order method. |
| title | A Parameter-Free Zeroth-Order Algorithm for Decentralized Stochastic Convex Optimization |
| topic | Optimization and Control |
| url | https://arxiv.org/abs/2603.15219 |