An information theoretic sparse kernel algorithm for online learning

作者：

Highlights：

• An information theoretic sparsification rule is proposed for kernel online learning.

• An adaptive learning rate is proposed based on the dead zone scheme.

• Guaranteed convergence analysis on kernel weight error vector is provided.

摘要

•An information theoretic sparsification rule is proposed for kernel online learning.•An adaptive learning rate is proposed based on the dead zone scheme.•Guaranteed convergence analysis on kernel weight error vector is provided.

论文关键词：Kernel methods,Information theoretic,Sparsification,Online learning,Mutual information

论文评审过程：Available online 24 January 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.01.010