A Physics-Inspired Algorithm for Bilevel optimization

This paper presents the application of a physics-inspired algorithm based on the center of mass concept, called Bilevel Centers Algorithm (BCA), to deal with bilevel optimization problems. The center of mass is adopted for creating new directions in the bilevel continuous search space considering the objective function values of a set of randomly-chosen solutions in a hierarchical optimization structure. The performance of this approach is assessed by using representative test functions for bilevel optimization. The obtained results are compared against the state-of-the algorithm BLEAQ. The results based on accuracy and number of evaluations are competitive and promising.