Unsupervised domain adaptation with progressive adaptation of subspaces

作者：

Highlights：

• explore a novel UDA method named Progressive Adaptation of Subspaces (PAS) without domain alignment for effectively alleviating the mode collapse in domain adaptation.

• provide an effective algorithm to implement PAS, which progressively anchors and leverages the target samples with reliable pseudo labels to refine the shared subspaces.

• demonstrate the effectiveness of PAS on UDA, especially on a more realistic and challenging scenario (i.e., partial domain adaptation).

摘要

•explore a novel UDA method named Progressive Adaptation of Subspaces (PAS) without domain alignment for effectively alleviating the mode collapse in domain adaptation.•provide an effective algorithm to implement PAS, which progressively anchors and leverages the target samples with reliable pseudo labels to refine the shared subspaces.•demonstrate the effectiveness of PAS on UDA, especially on a more realistic and challenging scenario (i.e., partial domain adaptation).

论文关键词：Unsupervised domain adaptation,Partial domain adaptation,Subspace learning,Pseudo label

论文评审过程：Received 29 August 2020, Revised 9 April 2022, Accepted 21 July 2022, Available online 23 July 2022, Version of Record 5 August 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108918