Gaussian Mixture Descriptors Learner
作者：
Highlights：
• A classification method based on minimum description length principle is proposed.
• The proposed method offered a good trade-off between complexity and predictive power.
• The proposed method is lightweight, multiclass, and offers incremental learning.
• Experiments were performed using sixteen datasets on two online learning scenarios.
• The proposed method outperformed most of the compared online learning methods.
摘要
•A classification method based on minimum description length principle is proposed.•The proposed method offered a good trade-off between complexity and predictive power.•The proposed method is lightweight, multiclass, and offers incremental learning.•Experiments were performed using sixteen datasets on two online learning scenarios.•The proposed method outperformed most of the compared online learning methods.
论文关键词：Minimum description length,Classification,Machine learning
论文评审过程：Received 29 March 2019, Revised 31 July 2019, Accepted 12 September 2019, Available online 14 September 2019, Version of Record 20 January 2020.
论文官网地址：https://doi.org/10.1016/j.knosys.2019.105039
