A Simple System of Evolutionary Operation Subject to Empirical Feedback
作者:
GeorgeE. P. Box,
期刊:
Technometrics
(Taylor Available online 1966)
卷期:
Volume 8,
issue 1
页码: 19-26
ISSN:0040-1706
年代: 1966
DOI:10.1080/00401706.1966.10490321
出版商: Taylor & Francis Group
数据来源: Taylor
摘要:
Experimentation can lead to progress throughempirical feedbackor throughscientific(technical)feedback; or more usually by some combination of both [Box and Jenkins, 19621. Empirical feedback is typical of the “try it and see” or Edisonian approach. Here information fed back from the experiment triggers a simple reaction like “accept a modification which increases yield.” By contrast scientific feedback occurs as an interaction between the data and the knowledge, experience, and imagination of the experimenter. It may result in reactions to data which are quite unpredictable by consideration of the data alone. An example would be the decision to vary in further experimentation certain specific factors not even considered in the previous stages of the work. In this paper an empirical feedback model for an elementary form of evolutionary operation is considered. From this investigation it is clear that the strategy necessary to achieve optimal empirical feedback might give very poor scientific feedback and vice versa. Because scientific feedback plays perhaps the most important part in many successful uses of evolutionary operation it would be inefficient in practice to adopt the optimal empirical feedback strategy.
点击下载:
PDF (581KB)
返 回