Research map / Label-efficient and robust learning / Domain adaptation and generalization
Universal Domain Adaptation through Self Supervision
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 |
|---|---|---|---|
| Apply DANCE rather than source-only training | ResNet50 universal domain adaptation · Closed, partial, open-set, and open-partial DA; Office, OfficeHome, VisDA | accuracy | Mixed |
| Add neighborhood clustering to the learned features (ablation) | DANCE · Open-set DA, OfficeHome (linear probe) | known accuracy: 42.4 -> 48.4; 43.4 -> 47.8; 69.9 -> 74.9; 38.6 -> 45.6; 37.0 -> 42.5; 65.1 -> 74.7 | Helped |
| Add entropy separation to neighborhood clustering (ablation) | NC features · Open-set DA, OfficeHome (linear probe) | known accuracy: 48.4 -> 49.1; 47.8 -> 48.7; 74.9 -> 74.9; 45.6 -> 46.4; 42.5 -> 43.0; 74.7 -> 74.1 | Mixed |
| Replace neighborhood clustering with jigsaw puzzle loss (ablation) | DANCE · Open-set DA, Office | accuracy: 92.0 -> 94.1 | Hurt |
| Replace neighborhood clustering with jigsaw puzzle loss (ablation) | DANCE · Closed DA, Office | accuracy: 82.5 -> 85.5 | Hurt |
Closest papers by meaning
- Attracting and Dispersing: A Simple Approach for Source-free Domain Adaptation (NeurIPS 2022 · similarity 0.91)
- Exploring Category-Agnostic Clusters for Open-Set Domain Adaptation (CVPR 2020 · similarity 0.90)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.89)
- Tackling Dimensional Collapse toward Comprehensive Universal Domain Adaptation (ICML 2025 · similarity 0.88)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.88)
- Dynamic Target Distribution Estimation for Source-Free Open-Set Domain Adaptation (AAAI 2025 · similarity 0.88)
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.