Machine Learning Powered Feasible Path Framework with Adaptive Sampling for Black-box Optimization

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Zixuan, Song, Xiaowei, Li, Jiaming, Zeng, Yujiao, Nie, Yaling, Zhu, Min, Lu, Dongyun, Zhang, Yibo, Xiao, Xin, Li, Jie
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915513594019840
author Zhang, Zixuan
Song, Xiaowei
Li, Jiaming
Zeng, Yujiao
Nie, Yaling
Zhu, Min
Lu, Dongyun
Zhang, Yibo
Xiao, Xin
Li, Jie
author_facet Zhang, Zixuan
Song, Xiaowei
Li, Jiaming
Zeng, Yujiao
Nie, Yaling
Zhu, Min
Lu, Dongyun
Zhang, Yibo
Xiao, Xin
Li, Jie
contents Black-box optimization (BBO) involves functions that are unknown, inexact and/or expensive-to-evaluate. Existing BBO algorithms face several challenges, including high computational cost from extensive evaluations, difficulty in handling complex constraints, lacking theoretical convergence guarantees and/or instability due to large solution quality variation. In this work, a machine learning-powered feasible path optimization framework (MLFP) is proposed for general BBO problems including complex constraints. An adaptive sampling strategy is first proposed to explore optimal regions and pre-filter potentially infeasible points to reduce evaluations. Machine learning algorithms are leveraged to develop surrogates of black-boxes. The feasible path algorithm is employed to accelerate theoretical convergence by updating independent variables rather than all. Computational studies demonstrate MLFP can rapidly and robustly converge around the KKT point, even training surrogates with small datasets. MLFP is superior to the state-of-the-art BBO algorithms, as it stably obtains the same or better solutions with fewer evaluations for benchmark examples.
format Preprint
id arxiv_https___arxiv_org_abs_2509_21077
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Machine Learning Powered Feasible Path Framework with Adaptive Sampling for Black-box Optimization
Zhang, Zixuan
Song, Xiaowei
Li, Jiaming
Zeng, Yujiao
Nie, Yaling
Zhu, Min
Lu, Dongyun
Zhang, Yibo
Xiao, Xin
Li, Jie
Optimization and Control
Black-box optimization (BBO) involves functions that are unknown, inexact and/or expensive-to-evaluate. Existing BBO algorithms face several challenges, including high computational cost from extensive evaluations, difficulty in handling complex constraints, lacking theoretical convergence guarantees and/or instability due to large solution quality variation. In this work, a machine learning-powered feasible path optimization framework (MLFP) is proposed for general BBO problems including complex constraints. An adaptive sampling strategy is first proposed to explore optimal regions and pre-filter potentially infeasible points to reduce evaluations. Machine learning algorithms are leveraged to develop surrogates of black-boxes. The feasible path algorithm is employed to accelerate theoretical convergence by updating independent variables rather than all. Computational studies demonstrate MLFP can rapidly and robustly converge around the KKT point, even training surrogates with small datasets. MLFP is superior to the state-of-the-art BBO algorithms, as it stably obtains the same or better solutions with fewer evaluations for benchmark examples.
title Machine Learning Powered Feasible Path Framework with Adaptive Sampling for Black-box Optimization
topic Optimization and Control
url https://arxiv.org/abs/2509.21077