Online transfer learning with multiple source domains for multi-class classification

作者：

Highlights：

• An online multi-source multiple classification transfer learning algorithm is proposed.

• The mistake bound of the proposed algorithm is derived.

• Experimental results illustrate that the proposed algorithm has good performance.

• The first study on online multi-source multiple classification transfer learning.

摘要

•An online multi-source multiple classification transfer learning algorithm is proposed.•The mistake bound of the proposed algorithm is derived.•Experimental results illustrate that the proposed algorithm has good performance.•The first study on online multi-source multiple classification transfer learning.

论文关键词：Online transfer learning,Transfer learning,Multi-class classification,Multiple source domains,Online learning

论文评审过程：Received 11 July 2019, Revised 13 September 2019, Accepted 22 October 2019, Available online 5 November 2019, Version of Record 7 February 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.105149