Research map / Label-efficient and robust learning / Domain adaptation and generalization
Zero-Shot Deep Domain Adaptation
Kuan-Chuan Peng, Ziyan Wu, Jan Ernst
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 ZDDA2 training with task-irrelevant dual-domain pairs | LeNet source-only classifier · MNIST->MNIST-M | accuracy (%): 39.04 -> 94.84 | Improved |
| add ZDDA2 with task-irrelevant pairs | LeNet source-only classifier · NIST->NIST-M | accuracy (%): 8.59 -> 34.25 | Improved |
| use a more semantically similar task-irrelevant dataset (EMNIST rather than Fashion-MNIST) | ZDDA2 · MNIST->MNIST-M | accuracy (%): 73.15 -> 94.84 | Improved |
| use MNIST (S 0.049) rather than EMNIST (0.178) as T-I data | ZDDA2 · Fashion-MNIST->Fashion-MNIST-M | accuracy (%): 51.55 -> 65.30 | Mixed |
| add ZDDA2 training with T-I RGB-D pairs, testing on RGB | GoogleNet · SUN RGB-D 10 scenes | accuracy (%): 31.58 -> 43.16 | Improved |
| add ZDDA2 training with T-I RGB-D pairs, testing on depth | GoogleNet · SUN RGB-D 10 scenes | accuracy (%): 51.84 -> 59.74 | Improved |
| add step 3 (joint network without L2 constraint, ZDDA3) to ZDDA2 (ablation) | ZDDA2 · SUN RGB-D 10 scenes (depth test) | accuracy (%): 59.74 -> 63.16 | Improved |
| add step 3 (ZDDA3) to ZDDA2 (ablation) | ZDDA2 · SUN RGB-D 10 scenes, AlexNet | accuracy (%): 51.05 -> 51.05 | No clear effect |
| add step 3 (ZDDA3) to ZDDA2 (ablation) | ZDDA2 · SUN RGB-D 10 scenes, 5-fold splits | mean accuracy (%): 56.89 -> 58.37 | Improved |
| add ZDDA2 on a different base architecture (SqueezeNet), depth test | SqueezeNet_v1.1 · SUN RGB-D 10 scenes | accuracy (%): 48.68 -> 56.32 | Improved |
| add ZDDA2 on AlexNet, depth test | AlexNet · SUN RGB-D 10 scenes | accuracy (%): 49.74 -> 51.05 | Improved |
Closest papers by meaning
- Zero-shot Deep Domain Adaptation with Common Representation Learning (TPAMI 2021 · similarity 0.97)
- Learning Across Tasks for Zero-Shot Domain Adaptation From a Single Source Domain (TPAMI 2021 · similarity 0.90)
- Collaborative Learning With Disentangled Features for Zero-Shot Domain Adaptation (ICCV 2021 · similarity 0.89)
- Unified Language-driven Zero-shot Domain Adaptation (CVPR 2024 · similarity 0.88)
- Source-Free Domain Adaptation with Frozen Multimodal Foundation Model (CVPR 2024 · similarity 0.87)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (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.
