Online transfer learning with partial feedback

作者：

Highlights：

• We propose a partial feedback online transfer learning (PFOTL) algorithm.

• We utilize transfer learning to improve learning in the partial feedback scenario.

• We theoretically analyze the mistake bound for the proposed PFOTL algorithm.

• Experiments on multiple datasets demonstrate the superior performance of PFOTL.

摘要

•We propose a partial feedback online transfer learning (PFOTL) algorithm.•We utilize transfer learning to improve learning in the partial feedback scenario.•We theoretically analyze the mistake bound for the proposed PFOTL algorithm.•Experiments on multiple datasets demonstrate the superior performance of PFOTL.

论文关键词：Online learning,Transfer learning,Partial feedback,Multi-class classification,Online transfer learning

论文评审过程：Received 20 December 2021, Revised 22 July 2022, Accepted 29 August 2022, Available online 6 September 2022, Version of Record 15 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118738