Research map / Label-efficient and robust learning / Domain adaptation and generalization
Efficient Inference under Label Shift in Unsupervised Domain Adaptation
Seong-ho Lee, Yanyuan Ma, Jiwei Zhao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace shift-dependent estimator with estimated density ratio efficient estimator (Efficient^) | Shift-dependent* · Simulation, E_q(Y) | MSE x100: 25.8428 -> 1.4096 | Improved |
| use efficient estimator Efficient^ instead of singly-flexible | Singly-flexible* · Simulation, E_q(Y) | MSE x100: 1.8039 -> 1.4096 | Improved |
| use efficient estimator Efficient^ instead of doubly-flexible | Doubly-flexible*⋆ · Simulation, E_q(Y) | MSE x100: 1.8189 -> 1.4096 | Improved |
| use efficient estimator Efficient^ instead of singly-flexible | Singly-flexible* · Simulation, var_q(Y) | MSE x100: 8.7310 -> 5.1272 | Improved |
| use efficient estimator Efficient^ instead of doubly-flexible | Doubly-flexible*⋆ · Simulation, var_q(Y) | MSE x100: 9.8121 -> 5.1272 | Improved |
| estimate rho(y) with self-improved estimator rho-hat instead of consistent rho-tilde (ablation) | Efficient · Simulation, density ratio rho(y) | variability of estimate | Improved |
| estimate rho(y) nonparametrically instead of misspecified working model rho*(y) | Shift-dependent* · Simulation, density ratio rho(y) | estimation accuracy | Improved |
| use Efficient^ instead of Singly-flexible | Singly-flexible* · Plankton counting | 95% CI width: 661.606 -> 433.062 | Improved |
| use Efficient^ instead of PPI | PPI* · Plankton counting | 95% CI width: 8466.000 -> 433.062 | Improved |
| use Efficient^ instead of Shift-dependent | Shift-dependent* · Plankton counting | 95% CI width: 578.271 -> 433.062 | Improved |
| use Efficient^ instead of Doubly-flexible | Doubly-flexible*⋆ · Szeged temperature forecast | 95% CI width: 1.976 -> 0.493 | Improved |
| use Efficient^ instead of Singly-flexible | Singly-flexible* · Szeged temperature forecast | 95% CI width: 0.629 -> 0.493 | Improved |
Closest papers by meaning
- Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs (NeurIPS 2026 · similarity 0.91)
- ELSA: Efficient Label Shift Adaptation through the Lens of Semiparametric Models (ICML 2023 · similarity 0.91)
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.89)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.88)
- Diffusion-Driven Progressive Target Manipulation for Source-Free Domain Adaptation (NeurIPS 2025 · similarity 0.88)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
