All algorithms that we are going to discuss are iterative. Let \(\xi _s\) be the design that was obtained after \((s-1)\) iteration. At the s-th iteration step, we will hopefully improve—as in all previous steps—the characteristics of the design. The new design, for instance, may be obtained according to \(\displaystyle \xi _{s+1} = (1 - \alpha _s) \xi _s + \alpha _s \xi \,; \) this means that we reduce the number of observations that are taken in accordance with design \(\xi _s\) and that we, instead, take some observations at points which correspond to design \(\xi \) . How shall we choose those points or the design \(\xi \) ?

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Numerical Techniques

  • Valerii V. Fedorov,
  • Peter Hackl

摘要

All algorithms that we are going to discuss are iterative. Let \(\xi _s\) be the design that was obtained after \((s-1)\) iteration. At the s-th iteration step, we will hopefully improve—as in all previous steps—the characteristics of the design. The new design, for instance, may be obtained according to \(\displaystyle \xi _{s+1} = (1 - \alpha _s) \xi _s + \alpha _s \xi \,; \) this means that we reduce the number of observations that are taken in accordance with design \(\xi _s\) and that we, instead, take some observations at points which correspond to design \(\xi \) . How shall we choose those points or the design \(\xi \) ?