Research map / Label-efficient and robust learning / Recognition and long-tail
Sum of Ranked Range Loss for Supervised Learning
Shu Hu, Yiming Ying, Xin Wang, Siwei Lyu
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 |
|---|---|---|---|
| use AoRR instead of average aggregate loss (logistic) | linear classifier · Monk | error rate (%): 20.46 -> 12.69 | Improved |
| use AoRR instead of AT_k aggregate loss (logistic) | linear classifier · Monk | error rate (%): 16.76 -> 12.69 | Improved |
| use AoRR instead of AT_k aggregate loss (hinge) | linear classifier · Monk | error rate (%): 17.04 -> 13.17 | Improved |
| use AoRR instead of AT_k aggregate loss (logistic) | linear classifier · Phoneme | error rate (%): 24.17 -> 21.95 | Improved |
| use AoRR instead of AT_k aggregate loss (logistic) | linear classifier · Titanic | error rate (%): 22.44 -> 21.69 | Improved |
| use AoRR instead of AT_k aggregate loss (logistic) | linear classifier · Australian | error rate (%): 11.7 -> 11.42 | Improved |
| use AoRR instead of AT_k aggregate loss (hinge) | linear classifier · Australian | error rate (%): 12.51 -> 12.5 | Improved |
| exclude the m largest individual losses (increase m) in AoRR | AoRR logistic · four real data sets | error rate | Mixed |
| use AoRR instead of AT_k on MNIST noise 0.2 | softmax loss · MNIST symmetric noise 0.2 | accuracy (%): 89.71 -> 92.42 | Improved |
| use AoRR instead of AT_k on MNIST noise 0.3 | softmax loss · MNIST symmetric noise 0.3 | accuracy (%): 88.73 -> 92.26 | Improved |
| use AoRR instead of AT_k on MNIST noise 0.4 | softmax loss · MNIST symmetric noise 0.4 | accuracy (%): 87.62 -> 91.87 | Improved |
| use AT_k instead of average loss on MNIST noisy labels | softmax loss · MNIST symmetric noise 0.2 | accuracy (%): 89.69 -> 89.71 | No clear effect |
Closest papers by meaning
- Learning by Minimizing the Sum of Ranked Range (NeurIPS 2020 · similarity 0.97)
- Rethinking and Reweighting the Univariate Losses for Multi-Label Ranking: Consistency and Generalization (NeurIPS 2021 · similarity 0.86)
- Rank-based Decomposable Losses in Machine Learning: A Survey (TPAMI 2023 · similarity 0.85)
- Bipartite Ranking From Multiple Labels: On Loss Versus Label Aggregation (ICML 2025 · similarity 0.85)
- Semi-supervised Multi-label Learning with Balanced Binary Angular Margin Loss (NeurIPS 2024 · similarity 0.83)
- Forming Auxiliary High-confident Instance-level Loss to Promote Learning from Label Proportions (CVPR 2025 · similarity 0.83)
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.
