Multi-objective bayesian optimization algorithm

Recently, signi cant development in the theory and design of competent genetic algorithms (GAs) has been achieved. By competent GA we mean genetic algorithms that can solve boundedly diAEcult problems quickly, accurately, and reliably. However, most of the existing competent GAs focus only on single-objective optimization although many real-world problems contain more than one objective. Independently of the development of competent genetic algorithms, a number of approaches to solve such multiobjective problems have been proposed. However, there has been little or no e ort to develop competent multiobjective operators that eAEciently identify, propagate, and combine important partial solutions of the problem at hand.