Research map / Label-efficient and robust learning / Domain adaptation and generalization
Collaborative and Adversarial Network for Unsupervised Domain Adaptation
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 |
|---|---|---|---|
| Add collaborative domain-informative learning at lower blocks and adversarial domain-invariant learning at higher blocks | ResNet50 domain adaptation · Office-31 average | accuracy: 80.9 -> 82.4 | Helped |
| Add collaborative domain-informative learning at lower blocks and adversarial domain-invariant learning at higher blocks | ResNet50 domain adaptation · ImageCLEF-DA average | accuracy: 85.8 -> 85.7 | No clear effect |
| Add iterative pseudo-label selection and reweighting using image and domain classifiers (ablation) | CAN · Office-31 average | accuracy: 82.4 -> 87.2 | Helped |
| Add iterative pseudo-label selection and reweighting using image and domain classifiers (ablation) | CAN · ImageCLEF-DA average | accuracy: 85.7 -> 87.4 | Helped |
| Add image-classifier target sample selection (ablation) | DANN · Office-31 A→W | accuracy: 79.3 -> 80.0 | Helped |
| Add image-classifier target sample selection (ablation) | CAN · Office-31 A→W | accuracy: 81.5 -> 86.6 | Helped |
| Use both image and domain classifiers for target sample selection (ablation) | DANN · Office-31 A→W | accuracy: 80.0 -> 85.3 | Helped |
| Use both image and domain classifiers for target sample selection and reweighting (ablation) | CAN · Office-31 A→W | accuracy: 86.6 -> 92.5 | Helped |
| Vary λ0 and λm loss-weight settings (ablation) | iCAN · Office-31 A→W | accuracy: 84.3 to 92.5 | Mixed |
| Vary pseudo-label sample-weight parameters α and β (ablation) | iCAN · Office-31 A→W | accuracy: 86.9 to 92.5 | Mixed |
Closest papers by meaning
- Adversarial Discriminative Domain Adaptation (CVPR 2017 · similarity 0.90)
- Unsupervised Domain Adaptation with Residual Transfer Networks (NeurIPS 2016 · similarity 0.90)
- Domain-Symmetric Networks for Adversarial Domain Adaptation (CVPR 2019 · similarity 0.90)
- Unsupervised Pixel-Level Domain Adaptation With Generative Adversarial Networks (CVPR 2017 · similarity 0.89)
- GCAN: Graph Convolutional Adversarial Network for Unsupervised Domain Adaptation (CVPR 2019 · similarity 0.88)
- Adversarial-Learned Loss for Domain Adaptation (AAAI 2020 · 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.