Properties and numerical testing of a parallel global optimization algorithm (Q715140): Difference between revisions

The global minimum of a real function \(f(x)\), \(x \in S\subseteq \mathbb{R}^n\) can be found using the algorithm proposed by \textit{M. Gaviano} et al. [J. Glob. Optim. 48, No. 1, 73--85 (2010; Zbl 1202.90214)]. The algorithm uses the optimal probability value to be set at each iteration so that by moving from a local minimum to a new one, the average number of function evaluations is minimal. In this paper, the authors consider the computational cost of the entire process of finding the global minimum and show that it exhibits the same general features as computational cost required to move from current local minimum to a new one. Moreover, they give the counterexample that the optimal values obtained using these two functions cannot agree. Finally, they present results of some numerical experiments written in Matlab and parallel toolboxes.