Improved multi-kernel classification machine with Nyström approximation technique

作者：

Highlights：

• Avoiding the problem of setting some initial parameters.

• Keeping a comparable space and computational complexities.

• Having a tighter generalization risk bound.

• Having a better recognition than NMKMHKS on average.

• Possessing an ability to deal with noise and practical images.

摘要

Highlights•Avoiding the problem of setting some initial parameters.•Keeping a comparable space and computational complexities.•Having a tighter generalization risk bound.•Having a better recognition than NMKMHKS on average.•Possessing an ability to deal with noise and practical images.

论文关键词：Multiple kernel learning,Nyström approximation,Generalization risk analysis,Pattern classification

论文评审过程：Received 10 June 2014, Revised 23 September 2014, Accepted 20 October 2014, Available online 31 October 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.10.029