Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning by Transferring from Unsupervised Universal Sources
Yu-Xiong Wang, Martial Hebert
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 |
|---|---|---|---|
| use unsupervised universal sources instead of naive clustering sources (ablation) | MT-SVM · Office Webcam Subset A | accuracy (%): 67.13 -> 74.83 | Improved |
| use unsupervised universal sources instead of naive clustering sources (ablation) | MT-SVM · Office Webcam Subset B | accuracy (%): 64.92 -> 74.19 | Improved |
| use unsupervised universal sources instead of supervised category sources (ablation) | MT-SVM · Office Webcam Subset B | accuracy (%): 70.30 -> 74.19 | Improved |
| add transfer from UUS vs target-only SVM (ablation) | MT-SVM · Office Webcam Subset B | accuracy (%): 63.34 -> 74.19 | Improved |
| increase gamma weight (m) from 0 to 0.7 in elastic net (ablation) | MT-SVM · SUN-397 one-shot | accuracy (%): 10.51 -> 15.54 | Improved |
| increase m from 0.7 to 0.9 (ablation) | MT-SVM · SUN-397 one-shot | accuracy (%): 15.54 -> 15.23 | No clear effect |
| generate UUS on PASCAL 2007 region proposals instead of ILSVRC (ablation) | MT-SVM · SUN-397 | accuracy (%) | Improved |
| elastic net MT-SVM instead of l2-only (EE-SVM) or l1-only (Multi-KT) transfer weights | MT-SVM · SUN-397 | accuracy (%) | Improved |
| elastic net MT-SVM instead of l2+l0 GreedyTL | MT-SVM · SUN-397 | accuracy (%) | Improved |
Closest papers by meaning
- Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfer (TPAMI 2021 · similarity 0.88)
- Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation (ICML 2020 · similarity 0.86)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.85)
- Let Go of Your Labels with Unsupervised Transfer (ICML 2024 · similarity 0.85)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.84)
- Towards Universal Dataset Distillation via Task-Driven Diffusion (CVPR 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.
