Logo eprints

Global and preference-based optimization using surrogate-based methods

Zhu, Mengjia (2025) Global and preference-based optimization using surrogate-based methods. Advisor: Bemporad, Prof. Alberto. pp. 187. [IMT PhD Thesis]

[img] Text (Doctoral thesis)
ZhuMengjia_Thesis_final version.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (3MB)


This thesis explores methodologies in black-box and preference-based optimization, addressing three key research questions. Firstly, it introduces a semi-automated calibration approach that eliminates the need for an explicit performance index by relying on human calibrator preferences. Secondly, the thesis delves into preference-based global optimization algorithms that address optimization problems where the analytic expression of the objective function is unknown and the optimization is subject to unknown constraints. The proposed algorithm, C-GLISp, extends the active preference learning framework to handle these unknown constraints. Lastly, the thesis tackles the challenge of optimization problems involving mixed variables and linear constraints. To address this challenge, we present a novel surrogate-based global optimization algorithm, named PWAS. The algorithm constructs a piecewise affine surrogate of the objective function over feasible samples and utilizes exploration functions to efficiently navigate the feasible domain using mixed-integer linear programming solvers. Additionally, a preference-based version of the algorithm, PWASp, is introduced to handle situations where only pairwise comparisons between samples are available instead of direct objective function evaluations. The efficiency and effectiveness of the proposed approaches are demonstrated via benchmark studies. Additionally, the practical applicability of PWAS is discussed via experimental design case stuides.

Item Type: IMT PhD Thesis
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
PhD Course: Computer science and systems engineering
Identification Number: https://doi.org/10.13118/imtlucca/e-theses/415
NBN Number: urn:nbn:it:imtlucca-30048
Date Deposited: 05 Jun 2024 13:14
URI: http://e-theses.imtlucca.it/id/eprint/415

Actions (login required, only for staff repository)

View Item View Item