期刊
MACHINE LEARNING
卷 110, 期 6, 页码 1429-1462出版社
SPRINGER
DOI: 10.1007/s10994-021-05992-x
关键词
Adaptive machine learning; Streaming data; Non-stationary data; Concept drift; Automated machine learning
This paper proposes the use of flexible adaptive mechanism deployment for automated development of adaptation strategies to address the issue in automated machine learning model development. Experimental results confirm the viability of these strategies, achieving better or comparable performance to custom adaptation strategies and repeated deployment of any single adaptive mechanism on 36 datasets.
Automation of machine learning model development is increasingly becoming an established research area. While automated model selection and automated data pre-processing have been studied in depth, there is, however, a gap concerning automated model adaptation strategies when multiple strategies are available. Manually developing an adaptation strategy can be time consuming and costly. In this paper we address this issue by proposing the use of flexible adaptive mechanism deployment for automated development of adaptation strategies. Experimental results after using the proposed strategies with five adaptive algorithms on 36 datasets confirm their viability. These strategies achieve better or comparable performance to the custom adaptation strategies and the repeated deployment of any single adaptive mechanism.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据