Research map / Label-efficient and robust learning / Recognition and long-tail
Posterior Re-calibration for Imbalanced Datasets
Junjiao Tian, Yen-Cheng Liu, Nathaniel Glaser, Yen-Chang Hsu, Zsolt Kira
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 imbalance calibration to CE-DRW (ablation) | ResNet32 on imbalanced CIFAR-10/100 · Imbalanced CIFAR-10/100, AVE | Top1 validation error: 33.62 -> 32.00 | Improved |
| Add imbalance calibration to cross-entropy (ablation) | InceptionV3 on iNaturalist2018 · iNaturalist2018 | Validation error: 35.35 -> 34.16 | Improved |
| Add imbalance calibration to CE-DRW (ablation) | ResNet50 on iNaturalist2018 · iNaturalist2018 | Validation error: 33.73 -> 32.06\pm 0.38 | Improved |
| Add imbalance calibration to cross-entropy (ablation) | DeepLab on Synthia · Synthia, in-distribution RGB | mIOU: 84.48 -> 84.71_{0.2} | Improved |
| Add imbalance calibration to UNO (ablation) | RGB-D segmentation on Synthia · Synthia, out-of-distribution weather, AVE | mACC: 82.74 -> 90.45 | Improved |
| Add imbalance calibration to UNO (ablation) | RGB-D segmentation on Synthia · Synthia, out-of-distribution weather, AVE | mIOU: 79.12 -> 78.55 | Hurt |
| Increase the prior-rebalancing weight toward the fully rebalanced posterior (ablation) | Imbalance calibration on Synthia · Synthia validation set | mACC | Improved |
| Increase the prior-rebalancing weight toward the fully rebalanced posterior (ablation) | Imbalance calibration on Synthia · Synthia validation set | mIOU | Hurt |
Closest papers by meaning
- Learning Imbalanced Data with Beneficial Label Noise (ICML 2025 · similarity 0.86)
- Balancing the Scales: A Theoretical and Algorithmic Framework for Learning from Imbalanced Data (ICML 2025 · similarity 0.86)
- Twice Class Bias Correction for Imbalanced Semi-supervised Learning (AAAI 2024 · similarity 0.85)
- Rebalancing Multi-Label Class-Incremental Learning (AAAI 2025 · similarity 0.85)
- Synthetic Tabular Data Generation for Imbalanced Classification: The Surprising Effectiveness of an Overlap Class (AAAI 2025 · similarity 0.84)
- Revive Re-weighting in Imbalanced Learning by Density Ratio Estimation (NeurIPS 2024 · similarity 0.84)
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.
