1: \begin{abstract}

2: We address an optimization problem where the cost function is the expectation of a random mapping.

3: To tackle the problem two approaches based on the approximation of the objective function by consensus-based particle optimization methods on the search space are developed. 

4: The resulting methods are mathematically analyzed using a mean-field approximation and their connection is established. 

5: Several numerical experiments show the validity of the proposed algorithms and investigate their rates of convergence.

6: \medskip

7: 

8: \noindent{\bf Keywords:} mean-field limit, particle swarm optimization, random optimization problems \\

9: 

10: \noindent{\bf MSC2000:} 82B40, 65K10, 60K35

11:  \end{abstract}

12: