Automated Adaptation Strategies for Stream Learning
This work addresses the gap in automated adaptation for stream learning, offering incremental improvements for researchers and practitioners in dynamic data environments.
The paper tackled the problem of automating model adaptation strategies in stream learning, where manually developing strategies is costly, and demonstrated that their proposed automated strategies achieve better or comparable performance to custom strategies and single mechanisms across 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.