Research map / Label-efficient and robust learning / Domain adaptation and generalization
Adversarial Discriminative Domain Adaptation
Check 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 |
|---|---|---|---|
| Apply ADDA adversarial adaptation instead of source-only classification | modified LeNet · MNIST → USPS | accuracy: 0.752 \pm 0.016 -> 0.894 \pm 0.002 | Helped |
| Apply ADDA adversarial adaptation instead of source-only classification | modified LeNet · USPS → MNIST | accuracy: 0.571 \pm 0.017 -> 0.901 \pm 0.008 | Helped |
| Apply ADDA adversarial adaptation instead of source-only classification | modified LeNet · SVHN → MNIST | accuracy: 0.601 \pm 0.011 -> 0.760 \pm 0.018 | Helped |
| Use ADDA rather than CoGAN | digit adaptation · MNIST → USPS | accuracy: 0.912 \pm 0.008 -> 0.894 \pm 0.002 | Hurt |
| Use ADDA rather than CoGAN | digit adaptation · USPS → MNIST | accuracy: 0.891 \pm 0.008 -> 0.901 \pm 0.008 | Helped |
| Apply ADDA adversarial adaptation instead of source-only classification | VGG-16 · NYUD RGB → depth | average per category accuracy: 0.139 -> 0.211 | Helped |
| Apply ADDA adversarial adaptation instead of source-only classification | VGG-16 · NYUD RGB → depth, counter | accuracy: 0.029 -> 0.447 | Helped |
| Apply ADDA adversarial adaptation instead of source-only classification | VGG-16 · NYUD RGB → depth, pillow | accuracy: 0.587 -> 0.297 | Hurt |
| Apply ADDA adversarial adaptation instead of source-only classification | VGG-16 · NYUD RGB → depth, night stand | accuracy: 0.039 -> 0.020 | Hurt |
| Apply ADDA adversarial adaptation instead of source-only classification | VGG-16 · NYUD RGB → depth | accuracy across object classes | Mixed |
Closest papers by meaning
- Unsupervised Pixel-Level Domain Adaptation With Generative Adversarial Networks (CVPR 2017 · similarity 0.93)
- f-Domain Adversarial Learning: Theory and Algorithms (ICML 2021 · similarity 0.92)
- Generate to Adapt: Aligning Domains Using Generative Adversarial Networks (CVPR 2018 · similarity 0.92)
- Multi-Adversarial Domain Adaptation (AAAI 2018 · similarity 0.91)
- Reusing the Task-Specific Classifier as a Discriminator: Discriminator-Free Adversarial Domain Adaptation (CVPR 2022 · similarity 0.91)
- Discriminative Adversarial Domain Adaptation (AAAI 2020 · similarity 0.91)
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.