Research map / Label-efficient and robust learning / Domain adaptation and generalization
Discriminative Adversarial 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 |
|---|---|---|---|
| use DADA instead of TADA | closed-set domain adaptation · Office-31 | average accuracy: 88.4 -> 89.0 | Helped |
| use DADA instead of ADR | closed-set domain adaptation · Syn2Real-C | mean accuracy: 74.8 -> 79.8 | Helped |
| use DADA-P instead of PADA | partial domain adaptation · Synthetic 12 → Real 6 | accuracy: 53.53 -> 69.06 | Helped |
| use DADA-O instead of AODA | open-set domain adaptation · Syn2Real-O, known-to-unknown ratio 1:1 | Known accuracy: 60 -> 66 | Helped |
| use DADA-O instead of AODA | open-set domain adaptation · Syn2Real-O, known-to-unknown ratio 1:1 | Mean accuracy: 62 -> 65 | Helped |
| use DADA-O instead of AODA | open-set domain adaptation · Syn2Real-O, known-to-unknown ratio 1:10 | Known accuracy: 51 -> 58 | Helped |
| use DADA-O instead of AODA | open-set domain adaptation · Syn2Real-O, known-to-unknown ratio 1:10 | Mean accuracy: 52 -> 57 | Helped |
| use DADA instead of MCD | digit domain adaptation · SVHN → MNIST; MNIST → USPS; USPS → MNIST | average accuracy: 94.8 -> 96.1 | Helped |
| add source and target discriminative adversarial losses (ablation) | DADA · Office-31 | average accuracy: 86.9 -> 88.1 | Helped |
| add entropy minimization loss (ablation) | DADA · Office-31 | average accuracy: 88.1 -> 89.0 | Helped |
| use the proposed target discriminative adversarial loss instead of the alternative target adversarial loss (ablation) | DADA · Office-31 | average accuracy: 87.9 -> 89.0 | Helped |
| use an integrated category and domain classifier instead of separate task and domain classifiers (ablation) | domain-adversarial adaptation · Office-31 | average accuracy: 82.5 -> 84.4 | Helped |
Closest papers by meaning
- Reusing the Task-Specific Classifier as a Discriminator: Discriminator-Free Adversarial Domain Adaptation (CVPR 2022 · similarity 0.91)
- Adversarial Discriminative Domain Adaptation (CVPR 2017 · similarity 0.91)
- A DIRT-T Approach to Unsupervised Domain Adaptation (ICLR 2018 · similarity 0.90)
- f-Domain Adversarial Learning: Theory and Algorithms (ICML 2021 · similarity 0.90)
- Domain-Symmetric Networks for Adversarial Domain Adaptation (CVPR 2019 · similarity 0.90)
- Multi-Adversarial Domain Adaptation (AAAI 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.