CatMADS: Mesh Adaptive Direct Search for constrained blackbox optimization with categorical variables

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Audet, Charles, Diouane, Youssef, Hallé-Hannan, Edward, Digabel, Sébastien Le, Tribes, Christophe
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912445851762688
author Audet, Charles
Diouane, Youssef
Hallé-Hannan, Edward
Digabel, Sébastien Le
Tribes, Christophe
author_facet Audet, Charles
Diouane, Youssef
Hallé-Hannan, Edward
Digabel, Sébastien Le
Tribes, Christophe
contents Solving optimization problems in which functions are blackboxes and variables involve different types poses significant theoretical and algorithmic challenges. Nevertheless, such settings frequently occur in simulation-based engineering design and machine learning. This paper extends the Mesh Adaptive Direct Search (MADS) algorithm to address mixed-variable problems with categorical, integer and continuous variables. MADS is a robust derivative-free optimization framework with a well-established convergence analysis for constrained quantitative problems. CatMADS generalizes MADS by incorporating categorical variables through distance-induced neighborhoods. A detailed convergence analysis of CatMADS is provided, with flexible choices balancing computational cost and local optimality strength. Four types of mixed-variable local minima are introduced, corresponding to progressively stronger notions of local optimality. CatMADS integrates the progressive barrier strategy for handling constraints, and ensures Clarke stationarity. An instance of \catmads employs cross-validation to construct problem-specific categorical distances. This instance is compared to state-of-the-art solvers on 32 mixed-variable problems, half of which are constrained. Data profiles show that CatMADS achieves the best results, demonstrating that the framework is empirically efficient in addition to having strong theoretical foundations.
format Preprint
id arxiv_https___arxiv_org_abs_2506_06937
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CatMADS: Mesh Adaptive Direct Search for constrained blackbox optimization with categorical variables
Audet, Charles
Diouane, Youssef
Hallé-Hannan, Edward
Digabel, Sébastien Le
Tribes, Christophe
Optimization and Control
90C11, 90C30, 90C56
Solving optimization problems in which functions are blackboxes and variables involve different types poses significant theoretical and algorithmic challenges. Nevertheless, such settings frequently occur in simulation-based engineering design and machine learning. This paper extends the Mesh Adaptive Direct Search (MADS) algorithm to address mixed-variable problems with categorical, integer and continuous variables. MADS is a robust derivative-free optimization framework with a well-established convergence analysis for constrained quantitative problems. CatMADS generalizes MADS by incorporating categorical variables through distance-induced neighborhoods. A detailed convergence analysis of CatMADS is provided, with flexible choices balancing computational cost and local optimality strength. Four types of mixed-variable local minima are introduced, corresponding to progressively stronger notions of local optimality. CatMADS integrates the progressive barrier strategy for handling constraints, and ensures Clarke stationarity. An instance of \catmads employs cross-validation to construct problem-specific categorical distances. This instance is compared to state-of-the-art solvers on 32 mixed-variable problems, half of which are constrained. Data profiles show that CatMADS achieves the best results, demonstrating that the framework is empirically efficient in addition to having strong theoretical foundations.
title CatMADS: Mesh Adaptive Direct Search for constrained blackbox optimization with categorical variables
topic Optimization and Control
90C11, 90C30, 90C56
url https://arxiv.org/abs/2506.06937