Transfer Learning under Group-Label Shift: A Semiparametric Exponential Tilting Approach
群组标签偏移下的迁移学习:半参数指数倾斜方法
We propose a new framework for binary classification in transfer learning settings where both covariate and label distributions may shift between source and target domains. Unlike traditional covariate shift or label shift assumptions, we introduce a group-label shift assumption that accommodates subpopulation imbalance and mitigates spurious correlations, thereby improving robustness to real-world distributional changes. To model the joint distribution difference, we adopt…