Discriminative and informative joint distribution adaptation for unsupervised domain adaptation

作者：

Highlights：

• A novel feature learning method DIJDA is proposed for unsupervised domain adaptation.

• A maximum margin criterion is used to preserve the separability of the samples.

• A row-sparsity regularization is adopted to identify the informative features.

• The validity of the proposed method is verified by extensive comparison experiments.

摘要

•A novel feature learning method DIJDA is proposed for unsupervised domain adaptation.•A maximum margin criterion is used to preserve the separability of the samples.•A row-sparsity regularization is adopted to identify the informative features.•The validity of the proposed method is verified by extensive comparison experiments.

论文关键词：Domain adaptation,Joint distribution adaptation,Maximum margin criterion,Row-sparsity

论文评审过程：Received 30 March 2020, Revised 22 July 2020, Accepted 8 August 2020, Available online 14 August 2020, Version of Record 20 August 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106394