Hyperparameter optimization of non-linear machine learning models using bi-level data-driven optimization

doi:10.1016/j.compchemeng.2026.109671

Hyperparameter optimization of non-linear machine learning models using bi-level data-driven optimization

2026-04-17

0

PRE

AI

DOI：10.1016/j.compchemeng.2026.109671

Original

Original request for help

Save

Abstract

En 中文

• Hyperparameter tuning of machine learning models is addressed using data-driven optimization. • Performance of 17 different data-driven optimization algorithms over 6 case studies is quantified. • The data-driven bi-level approach outperforms grid search, random search, and Bayesian optimization. • Results show enhanced predictive performance on the testing sets for regression and classification problems.

Keywords:

hyperparameter optimization

data-driven optimization

bi-level approach

machine learning models

predictive performance

Hyperparameter optimization of non-linear machine learning models using bi-level data-driven optimization

Abstract

Journal

Organization