Research map / Label-efficient and robust learning / Domain adaptation and generalization
Exploring Category-Agnostic Clusters for Open-Set 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 |
|---|---|---|---|
| Add conditional entropy loss to Self-Ensembling (ablation) | SE on VisDA open-set · VisDA open-set | Mean accuracy: 65.2 -> 66.3 | Helped |
| Add KL cluster-distribution loss to SE + CE (ablation) | SE-CC on VisDA open-set · VisDA open-set | Mean accuracy: 66.3 -> 69.3 | Helped |
| Add mutual information maximization to SE + CE + KL (ablation) | SE-CC on VisDA open-set · VisDA open-set | Mean accuracy: 69.3 -> 70.5 | Helped |
| Replace KL divergence with L1 distance for cluster-distribution mismatch (ablation) | SE-CC on VisDA open-set · VisDA open-set | Mean accuracy: 70.5 -> 68.7 | Hurt |
| Replace KL divergence with L2 distance for cluster-distribution mismatch (ablation) | SE-CC on VisDA open-set · VisDA open-set | Mean accuracy: 70.5 -> 68.4 | Hurt |
| Estimate mutual information using classification output only rather than both task outputs (ablation) | SE-CC on VisDA open-set · VisDA open-set | Mean accuracy: 70.5 -> 69.4 | Hurt |
| Estimate mutual information using clustering output only rather than both task outputs (ablation) | SE-CC on VisDA open-set · VisDA open-set | Mean accuracy: 70.5 -> 70.1 | Hurt |
| Use category-agnostic clusters and their structure-preservation guidance in Self-Ensembling | SE on Office open-set · Office open-set (Avg OS) | OS accuracy: 73.5 -> 90.4 | Helped |
| Use category-agnostic clusters and their structure-preservation guidance in Self-Ensembling | SE on VisDA open-set · VisDA open-set | Mean accuracy: 65.2 -> 70.5 | Helped |
| Refresh clusters periodically using learned features | SE-CC · Office and VisDA domain adaptation | performance | No clear effect |
Closest papers by meaning
- Dynamic Target Distribution Estimation for Source-Free Open-Set Domain Adaptation (AAAI 2025 · similarity 0.91)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.88)
- Open-Set Heterogeneous Domain Adaptation: Theoretical Analysis and Algorithm (AAAI 2025 · similarity 0.88)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.87)
- Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment (CVPR 2025 · similarity 0.87)
- Target Semantics Clustering via Text Representations for Robust Universal Domain Adaptation (AAAI 2025 · similarity 0.87)
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.