Imbalanced Open Set Domain Adaptation via Moving-threshold Estimation and Gradual Alignment

1 code implementation • 8 Mar 2023 • Jinghan Ru, Jun Tian, Zhekai Du, Chengwei Xiao, Jingjing Li, Heng Tao Shen

To alleviate the negative effects raised by label shift in OSDA, we propose Open-set Moving-threshold Estimation and Gradual Alignment (OMEGA) - a novel architecture that improves existing OSDA methods on class-imbalanced data.

Transfer Learning Unsupervised Domain Adaptation