Research map / Label-efficient and robust learning / Recognition and long-tail
PUe: Biased Positive-Unlabeled Learning Enhancement by Causal Inference
Xutao Wang, Hanting Chen, Tianyu Guo, Yunhe Wang
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 PUe propensity weighting to Dist-PU | Dist-PU · MNIST | ACC: 89.79 -> 92.57 | Improved |
| add PUe propensity weighting to Dist-PU | Dist-PU · CIFAR-10 | ACC: 84.45 -> 86.97 | Improved |
| add PUe propensity weighting to Dist-PU | Dist-PU · Alzheimer | ACC: 69.74 -> 70.51 | Improved |
| add PUe propensity weighting to nnPU | nnPU · MNIST | ACC: 87.81 -> 92.45 | Improved |
| add PUe propensity weighting to uPU | uPU · CIFAR-10 | F1: 63.01 -> 70.73 | Improved |
| add PUe propensity weighting to PUbN | PUbN · CIFAR-10 | ACC: 84.54 -> 85.66 | Improved |
| remove normalization of propensity weights (ablation) | nnPUe · MNIST | ACC: 90.95 -> 92.45 | Improved |
| remove alpha_e regularization term (ablation) | Dist-PUe · CIFAR-10 labeled dist [.08,.86,.02,.04] | ACC: 86.39 -> 88.15 | Improved |
| remove alpha_e regularization term (ablation) | Dist-PUe · CIFAR-10 labeled dist [.10,.10,.30,.50] | ACC: 90.40 -> 91.50 | Improved |
| use PUe under nearly uniform labeled distribution (ablation) | Dist-PUe · CIFAR-10 labeled dist [.25,.25,.25,.25] | ACC: 91.88 -> 91.79 | No clear effect |
| use estimated instead of known propensity scores | Dist-PUe · MNIST | ACC: 92.57 -> 91.73 | Hurt |
| use estimated instead of known propensity scores | nnPUe · MNIST | ACC: 92.45 -> 92.94 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Dist-PU — Dist-PU: Positive-Unlabeled Learning From a Label Distribution Perspective (CVPR 2022) · better on 12 of 13 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms (ICLR 2026 · similarity 0.91)
- Recovering the Propensity Score from Biased Positive Unlabeled Data (AAAI 2022 · similarity 0.90)
- Learning from positive and unlabeled examples -Finite size sample bounds (NeurIPS 2025 · similarity 0.89)
- Dist-PU: Positive-Unlabeled Learning From a Label Distribution Perspective (CVPR 2022 · similarity 0.89)
- Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning (NeurIPS 2025 · similarity 0.89)
- Positive and Unlabeled Learning with Controlled Probability Boundary Fence (ICML 2024 · 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.
