Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain2Vec: Domain Embedding for Unsupervised Domain Adaptation
Xingchao Peng, Yichen Li, Kate Saenko
Read the paperCheck your idea against 137,217 papers
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 Gram matrix information (ablation) | Domain2Vec · TinyDA MSDA ablation avg | accuracy (%): 43.5 -> 44.9 | Improved |
| add mutual information minimization (ablation) | Domain2Vec · TinyDA MSDA ablation avg | accuracy (%): 44.0 -> 44.9 | Improved |
| add Gram matrix information (ablation) | Domain2Vec · DomainBank openset DA avg | accuracy (%): 72.2 -> 73.8 | Improved |
| add mutual information minimization (ablation) | Domain2Vec · DomainBank openset DA avg | accuracy (%): 72.9 -> 73.8 | Improved |
| add Gram matrix information (ablation) | Domain2Vec · DomainBank partial DA avg | accuracy (%): 63.7 -> 65.5 | Improved |
| add mutual information minimization (ablation) | Domain2Vec · DomainBank partial DA avg | accuracy (%): 63.5 -> 65.5 | Improved |
| weight moment-matching MSDA loss by Domain2Vec domain distance (alpha) | M3SDA · TinyDA MSDA avg (GS CIFAR target) | accuracy (%): 46.8 -> 48.5 | Improved |
| weight adversarial MSDA loss by Domain2Vec domain distance (beta) | DCTN-style adversarial MSDA · TinyDA MSDA avg (GS CIFAR target) | accuracy (%): 48.1 -> 49.7 | Improved |
| weight partial adversarial alignment by Domain2Vec domain distance | PADA · DomainBank partial DA avg | accuracy (%): 64.6 -> 65.5 | Improved |
| use Domain2Vec domain distance in openset DA | AODA · DomainBank openset DA avg | accuracy (%): 71.3 -> 73.8 | Improved |
| add Gram matrix information (single target MNIST->USPS) (ablation) | Domain2Vec · TinyDA MNIST->USPS | accuracy (%): 28.5 -> 28.2 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AutoDIAL — AutoDIAL: Automatic DomaIn Alignment Layers (ICCV 2017) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Unsupervised Domain Adaptation With Similarity Learning (CVPR 2018 · similarity 0.91)
- PUnDA: Probabilistic Unsupervised Domain Adaptation for Knowledge Transfer Across Visual Categories (ICCV 2017 · similarity 0.90)
- Maximum Structural Generation Discrepancy for Unsupervised Domain Adaptation (TPAMI 2022 · similarity 0.90)
- Domain-Agnostic Mutual Prompting for Unsupervised Domain Adaptation (CVPR 2024 · similarity 0.89)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.89)
- Preserving Clusters in Prompt Learning for Unsupervised Domain Adaptation (CVPR 2025 · similarity 0.89)
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.
