Research map / Label-efficient and robust learning / Domain adaptation and generalization
Transfer Learning via Minimizing the Performance Gap Between Domains
Boyu Wang, Jorge Mendez, Mingbo Cai, Eric Eaton
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 |
|---|---|---|---|
| add performance-gap-aware weight updates and weight clipping (gapBoost) over TransferBoost | gapBoost · 20 Newsgroups | error rate (%): 8.83 -> 7.68 (comp vs sci) | Improved |
| gapBoost vs TransferBoost on comp vs talk | gapBoost · 20 Newsgroups | error rate (%): 6.45 -> 7.10 | Hurt |
| gapBoost vs TransferBoost on Office-Caltech | gapBoost · Office-Caltech | error rate (%): 27.86 -> 27.06 (A->C); 44.88 -> 41.74 (W->D) | Mixed |
| use gapBoost instead of target-only AdaBoost | gapBoost · Office-Caltech | error rate (%): 43.87 -> 27.06 (A->C) | Improved |
| increase fraction of target training data (1% to 80%) | gapBoost vs baselines · 20 Newsgroups | error rate (%) | Improved |
| vary rho_S (exp(rho_S) 0.1 to 0.9) with rho_T=0 (ablation) | gapBoost · 20 Newsgroups | error rate (%) | Mixed |
| up-weight target instances when auxiliary learners agree (rho_T >= 0, large rho_S) (ablation) | gapBoost · 20 Newsgroups | error rate (%) | Improved |
Closest papers by meaning
- A Difference Standardization Method for Mutual Transfer Learning (ICML 2022 · similarity 0.89)
- Minimax Lower Bounds for Transfer Learning with Linear and One-hidden Layer Neural Networks (NeurIPS 2020 · similarity 0.89)
- Gap Minimization for Knowledge Sharing and Transfer (JMLR 2023 · similarity 0.88)
- Wasserstein Transfer Learning (NeurIPS 2025 · similarity 0.87)
- A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning (NeurIPS 2025 · similarity 0.86)
- Theoretical Performance Guarantees for Partial Domain Adaptation via Partial Optimal Transport (ICML 2025 · similarity 0.86)
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.
