报告人：白富生 教授 （重庆师范大学）
报告摘要：In this talk, we present some stochastic adaptive stochastic algorithms using radial basis function models for global optimization of costly black-box functions. The exploration radii in local searches are generated adaptively. Each iteration point is selected from some randomly generated trial points according to certain criteria. Restarting strategies are adopted to build the restarting versions of these algorithms. The performance of the presented algorithms and their restarting versions are tested on some standard numerical examples. The numerical results indicate that these stochastic algorithms are very effective.