Online learning using projections onto shrinkage closed balls for adaptive brain-computer interface

作者：

Highlights：

• A shrinkage APSM algorithm was proposed and evaluated for BCI online learning.

• Limit point of shrinkage APSM was proved to approach to the least norm optimal solution.

• Better performance was obtained by shrinkage APSM, than general APSM, ISVM, PA.

• Tuning of the proposed method was shown to be easy.

摘要

•A shrinkage APSM algorithm was proposed and evaluated for BCI online learning.•Limit point of shrinkage APSM was proved to approach to the least norm optimal solution.•Better performance was obtained by shrinkage APSM, than general APSM, ISVM, PA.•Tuning of the proposed method was shown to be easy.

论文关键词：Online learning,Projections,Wearable/portable brain computer interface,Electroencephalography,Event-related potential,Biometrics

论文评审过程：Received 21 November 2018, Revised 19 July 2019, Accepted 18 August 2019, Available online 19 August 2019, Version of Record 26 August 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107017