Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Domain Adaptation with Residual Transfer Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 DAN's multiple MMD penalties with single tensor MMD feature fusion (ablation) | AlexNet domain adaptation · Office-31 Avg | accuracy: 71.7 -> 72.1 | Helped |
| Replace DAN's multiple MMD penalties with single tensor MMD feature fusion (ablation) | AlexNet domain adaptation · Office-Caltech Avg | accuracy: 90.1 -> 90.0 | No clear effect |
| Add target entropy minimization (ablation) | RTN (mmd) · Office-31 Avg | accuracy: 72.1 -> 72.9 | Helped |
| Add target entropy minimization (ablation) | RTN (mmd) · Office-Caltech Avg | accuracy: 90.0 -> 92.6 | Helped |
| Add residual classifier adaptation (ablation) | RTN (mmd+ent) · Office-31 Avg | accuracy: 72.9 -> 73.7 | Helped |
| Add residual classifier adaptation (ablation) | RTN (mmd+ent) · Office-Caltech Avg | accuracy: 92.6 -> 93.4 | Helped |
| Use the full RTN instead of DAN | AlexNet domain adaptation · Office-31 Avg | accuracy: 71.7 -> 73.7 | Helped |
| Use the full RTN instead of DAN | AlexNet domain adaptation · Office-Caltech Avg | accuracy: 90.1 -> 93.4 | Helped |
Closest papers by meaning
- Residual Parameter Transfer for Deep Domain Adaptation (CVPR 2018 · similarity 0.93)
- Learning Transferrable Representations for Unsupervised Domain Adaptation (NeurIPS 2016 · similarity 0.92)
- Transferable Adversarial Training: A General Approach to Adapting Deep Classifiers (ICML 2019 · similarity 0.91)
- Unsupervised Domain Adaptation With Similarity Learning (CVPR 2018 · similarity 0.91)
- Deep Asymmetric Transfer Network for Unbalanced Domain Adaptation (AAAI 2018 · similarity 0.91)
- Co-regularized Alignment for Unsupervised Domain Adaptation (NeurIPS 2018 · similarity 0.90)
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.