Discrepant collaborative training by Sinkhorn divergences

作者：

Highlights：

• An improved Co-training framework by applying Sinkhorn divergence module to it

• An empirical study of effectiveness of well-known variants of Sinkhorn divergence

• An extensive set of experiments demonstrating the advantage of our module

摘要

•An improved Co-training framework by applying Sinkhorn divergence module to it•An empirical study of effectiveness of well-known variants of Sinkhorn divergence•An extensive set of experiments demonstrating the advantage of our module

论文关键词：Co-training,Noisy labels,Weak-supervised learning

论文评审过程：Received 10 January 2021, Revised 12 May 2021, Accepted 13 May 2021, Available online 21 May 2021, Version of Record 16 June 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104213