1: \begin{abstract}

2: In the last two decades, significant effort has been made to solve computationally expensive optimization problems using surrogate models.

3: Regardless of whether surrogates are the primary drivers of an algorithm or improve the convergence of an existing method, most proposed concepts are rather specific and not very generalizable. Some important considerations are the selection of a baseline optimization algorithm, a suitable surrogate methodology, and the surrogate's involvement in the overall algorithm design.

4: This paper proposes a family of probabilistic surrogate-assisted algorithms (PSA), demonstrating its applicability to a broad category of single-objective optimization algorithms.

5: The concept introduces knowledge from a surrogate into an existing algorithm through a tournament-based procedure and continuing optimization on the surrogate's predictions. The surrogate's involvement is determined by updating a replacement probability based on the accuracy from past iterations.

6: A study of four well-known population-based optimization algorithms with and without the proposed probabilistic surrogate-assistance indicates its usefulness in achieving a better convergence. The concept should provide a generic and comprehensive use of surrogates within an optimization algorithm and pave the way for new surrogate-assisted algorithms dealing with challenges in less frequently addressed computationally expensive functions, such as different variable types, large dimensional problems, multiple objectives, and constraints.

7: \end{abstract}

8: