首页> 外文期刊>International Journal of Statistics and Probability >Simulation-Based Optimization for Convex Functions Over Discrete Sets
【24h】

Simulation-Based Optimization for Convex Functions Over Discrete Sets

机译:基于模拟的离散集凸函数的优化

获取原文
           

摘要

We propose a new iterative algorithm for finding a minimum point of f? : X ? R d → R, when f? is known to be convex, but only noisy observations of f?(x) are available at x ∈ X for a finite set X. At each iteration of the proposed algorithm, we estimate the probability of each point x ∈ X being a minimum point of f? using the fact that f? is convex, and sample r points from X according to these probabilities. We then make observations at the sampled points and use these observations to update the probability of each point x ∈ X being a minimum point of f?. Therefore, the proposed algorithm not only estimates the minimum point of f? but also provides the probability of each point in X being a minimum point of f?. Numerical results indicate the proposed algorithm converges to a minimum point of f? as the number of iterations increases and shows fast convergence, especially in the early stage of the iterations.
机译:我们提出了一种新的迭代算法来查找F的最小点? : X ? r d→r,当f? 已知是凸出的,但是只有F?(x)的噪声观察在x∈x时可用于有限组x。在所提出的算法的每次迭代时,我们估计每个点x∈x的概率是最小点的概率 离开? 使用f的事实? 根据这些概率凸,并根据这些概率对X分数。 然后,我们在采样点进行观察,并使用这些观察来更新每个点x∈X的概率是f的最小点。 因此,所提出的算法不仅估计F的最小点? 但也提供x中每个点的概率是f的最小点? 数值结果表明所提出的算法会聚到F的最小点? 随着迭代的数量增加并显示出快速收敛,特别是在迭代的早期阶段。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号