A federated data-driven evolutionary algorithm

Data-driven evolutionary optimization has witnessed great success in solving complex real-world problems. However, existing data-driven algorithms require that all data are centrally stored, which is not always practical and may be vulnerable to privacy leakage security threats if the must collected from different devices. To address above issue, this paper proposes a federated framework able perform driven when distributed on multiple On basis of learning, sorted model aggregation method developed for aggregating local surrogates based radial-basis-function networks. In addition, surrogate management strategy suggested by designing an acquisition function takes into account information both global models. Empirical studies set widely used benchmark functions presence various distributions demonstrate effectiveness proposed framework.