Session:

OBUPM - Optimization by Building and Using Probabilistic Models

Title:

Practical, Robust, and Scalable Black-Box Optimization with BOA and hBOA

Authors:

Martin Pelikan

Abstract:

To design practical black-box optimizers, one of the primary goals is to minimize the amount of work that must be done by the user while ensuring that a high-quality solution will be found quickly and reliably. This paper shows that probabilistic model-building genetic algorithms (PMBGAs) provide a great framework for designing practical and powerful black-box optimizers. The paper focuses on two algorithms that are among the most powerful PMBGAs: The Bayesian optimization algorithm (BOA) and the hierarchical BOA (hBOA).