Research map / Label-efficient and robust learning / Domain adaptation and generalization
Improving Domain-Specific Classification by Collaborative Learning with Adaptation 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 |
|---|---|---|---|
| add collaborative learning with target-specific network (no CR) (ablation) | adaptation network · MNIST→SVHN | accuracy (%): 71.6±0.5 -> 83.9±0.6 | Helped |
| add collaborative learning with target-specific network (no CR) (ablation) | adaptation network · MNIST→USPS | accuracy (%): 97.7±0.1 -> 97.7±0.1 | No clear effect |
| add collaborative learning (no CR) (ablation) | adaptation network · VisDA-2017 mean | accuracy (%): 67.5 -> 71.9 | Helped |
| add clustering regularization (ablation) | target-specific network · MNIST→SVHN | accuracy (%): 83.9±0.6 -> 84.9±0.4 | Helped |
| add clustering regularization (ablation) | target-specific network · STL→CIFAR | accuracy (%): 74.5±0.9 -> 75.4±0.2 | Helped |
| add clustering regularization (ablation) | target-specific network · VisDA-2017 mean | accuracy (%): 71.9 -> 74.3 | Helped |
| add clustering regularization (ablation) | target-specific network · VisDA-2017 truck class | accuracy (%): 42.6 -> 28.4 | Hurt |
| add clustering regularization (ablation) | target-specific network · SVHN→MNIST | accuracy (%): 99.0±0.3 -> 99.3±0.1 | No clear effect |
| use target-specific network instead of adaptation network for prediction | collaborative model · VisDA-2017 classes | accuracy (%) | Helped |
| full model vs DIRT-T | state-of-the-art · MNIST→SVHN | accuracy (%): 76.5 -> 84.9 | Helped |
| full model vs DIRT-T | state-of-the-art · SVHN→MNIST | accuracy (%): 99.4 -> 99.3 | No clear effect |
| full model vs Self-Ensembling | state-of-the-art · STL→CIFAR | accuracy (%): 69.2±0.4 -> 75.4±0.2 | Helped |
Closest papers by meaning
- Multi-Target Domain Adaptation With Collaborative Consistency Learning (CVPR 2021 · similarity 0.90)
- Co-regularized Alignment for Unsupervised Domain Adaptation (NeurIPS 2018 · similarity 0.89)
- Preserving Clusters in Prompt Learning for Unsupervised Domain Adaptation (CVPR 2025 · similarity 0.88)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.88)
- Open-Set Graph Domain Adaptation via Separate Domain Alignment (AAAI 2024 · similarity 0.87)
- Collaborative Semantic Consistency Alignment for Blended-Target Domain Adaptation (AAAI 2025 · similarity 0.87)
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.