Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Accuracy Estimation of Deep Visual Models using Domain-Adaptive Adversarial Perturbation without Source Samples
JoonHo Lee, Jae Oh Woo, Hankyu Moon, Kwonho Lee
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 adjusted predictive uncertainty scaling (C_adj_unc) to uniform epsilon scaling (ablation) | SF-DAP (ADV) · Office-31 | MAE (%): 5.76 -> 4.43 | Improved |
| uniformly scale epsilon by class complexity and density for all instances (ablation) | SF-DAP (ADV) · Office-31 | MAE (%): 4.72 -> 5.76 | Hurt |
| add class complexity factor C_cls with C_adj_unc (ablation) | SF-DAP (AAP) · Office-31 | MAE (%): 4.70 -> 3.05 | Improved |
| add data density factor C_den on top of C_cls and C_adj_unc (ablation) | SF-DAP (AAP) · Office-31 | MAE (%): 3.05 -> 2.51 | Improved |
| replace ADV with adaptive adversarial perturbation (AAP) (ablation) | SF-DAP · Overall macro average (63 settings) | MAE (%): 4.86 -> 2.95 | Improved |
| apply AAP directly to source model without source-free UDA (ablation) | SF-DAP (AAP) · Digits and Office-31 | average MAE: 3.0% increase in average MAE | Hurt |
| use SF-DAP (AAP) instead of RM under synthetic corruption shift | CIFAR-10-C / CIFAR-100-C · CIFAR-10-C | MAE (%): RM 2.17 vs AAP 3.06 | Hurt |
| use source-free SF-DAP (AAP) instead of source-access baselines on natural shift | Digits, Office-31, VisDA · VisDA | MAE (%): RM 4.41 -> AAP 1.73 | Improved |
| use PAFA instead of SHOT-IM for source-free UDA (ablation) | source-free UDA · M->S | target accuracy (%): 11.0 -> 83.5 | Improved |
| use PAFA instead of FAUST (ablation) | source-free UDA · UDA average | target accuracy (%): 94.6 -> 93.8 | No clear effect |
Closest papers by meaning
- Toward Improving Robustness and Accuracy in Unsupervised Domain Adaptation (AAAI 2025 · similarity 0.90)
- Adaptive Adversarial Network for Source-Free Domain Adaptation (ICCV 2021 · similarity 0.88)
- Bridging Theory and Practice in Source-Free Domain Adaptation via Adversarial Proxy Perturbation (ECCV 2026 · similarity 0.87)
- TITAN: Query-Token based Domain Adaptive Adversarial Learning (ICCV 2025 · similarity 0.87)
- Can We Evaluate Domain Adaptation Models Without Target-Domain Labels? (ICLR 2024 · similarity 0.87)
- Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: the Benefit of Target Expectation Maximization (ECCV 2018 · 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.
