In this paper, a hardware implementation of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithm is presented. This algorithm is based on the adaptation of the covariance matrix, initially it is focused on a region of the particular search space and subsequently, it moves or grows along the search space, as appropriate to find the optimum value. The experimental results reveal that this implementation will be very useful to resolve numeric optimization issues in an embedded system.
L. Urbina, Carlos A. Duchanoy, Marco A. Moreno-Armendáriz
Res. Comput. Sci.