Research map / Label-efficient and robust learning / Domain adaptation and generalization
Mixture Weight Estimation and Model Prediction in Multi-source Multi-target Domain Adaptation
Yuyang Deng, Ilja Kuzborskij, Mehrdad Mahdavi
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 |
|---|---|---|---|
| weighted ERM with learned mixture weights instead of average-weight ERM | Our method · MNIST target Group 1 | accuracy: 69.9 -> 80.0 | Improved |
| weighted ERM with learned weights instead of average ERM | Our method · MNIST target Group 2 | accuracy: 40.0 -> 69.9 | Improved |
| weighted ERM with learned weights instead of average ERM | Our method · MNIST target Group 3 | accuracy: 34.9 -> 55.0 | Improved |
| weighted ERM with learned weights instead of average ERM | Our method · MNIST target mix of Group 1 and 2 | accuracy: 59.9 -> 65.0 | Improved |
| use source data with learned weights instead of training only on target data | Our method · MNIST target Group 2 | accuracy: 55.0 -> 69.9 | Improved |
| use source data with learned weights instead of training only on target data | Our method · MNIST target Group 1 | accuracy: 69.9 -> 80.0 | Improved |
| pure target training instead of average ERM | Pure target training · MNIST target mix of Group 1 and 2 | accuracy: 59.9 -> 55.0 | Hurt |
Closest papers by meaning
- Principled Model Routing for Unknown Mixtures of Source Domains (NeurIPS 2025 · similarity 0.90)
- Scaling Laws for Optimal Data Mixtures (NeurIPS 2025 · similarity 0.87)
- Algorithms and Theory for Multiple-Source Adaptation (NeurIPS 2018 · similarity 0.87)
- A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning (NeurIPS 2025 · similarity 0.85)
- Revisiting the Domain Shift and Sample Uncertainty in Multi-source Active Domain Transfer (CVPR 2024 · similarity 0.84)
- GRAPE: Optimize Data Mixture for Group Robust Multi-target Adaptive Pretraining (NeurIPS 2025 · similarity 0.84)
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.
