mlsanityData to 7 Oct 2026

Classification from Pairwise Similarity and Unlabeled Data

arXiv 1802.04381 ↗First posted 12 Feb 2018cs.LGcited by 36

Not on the map. Why?

Abstract (arXiv)

Supervised learning needs a huge amount of labeled data, which can be a big bottleneck under the situation where there is a privacy concern or labeling cost is high. To overcome this problem, we propose a new weakly-supervised learning setting where only similar (S) data pairs (two examples belong to the same class) and unlabeled (U) data points are needed instead of fully labeled data, which is called SU classification. We show that an unbiased estimator of the classification risk can be obtained only from SU data, and the estimation error of its empirical risk minimizer achieves the optimal parametric convergence rate. Finally, we demonstrate the effectiveness of the proposed method through experiments.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). First posted on arXiv 12 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 9; 2022: 4; 2023: 8; 2024: 3; 2025: 7; 2026: 5.

36 papers.

Sort:

  1. 2021Understanding Negative Samples in Instance Discriminative Self-supervised Representation Learningcited by 47
  2. 2021On the Surrogate Gap between Contrastive and Supervised Lossescited by 25
  3. 2022Is the Performance of My Deep Network Too Good to Be True? A Direct Approach to Estimating the Bayes Error in Binary Classificationcited by 24
  4. 2021Learning from Similarity-Confidence Datacited by 19
  5. 2023Binary Classification with Confidence Differencecited by 13
  6. 2021Binary Classification from Multiple Unlabeled Datasets via Surrogate Set Classificationcited by 12
  7. 2023EEGMatch: Learning with Incomplete Labels for Semi-Supervised EEG-based Cross-Subject Emotion Recognitioncited by 10
  8. 2021Learning with Proper Partial Labelscited by 8
  9. 2021Positive-Unlabeled Classification under Class-Prior Shift: A Prior-invariant Approach Based on Density Ratio Estimationcited by 7
  10. 2024A General Framework for Learning from Weak Supervisioncited by 7
  11. 2022Empirical Evaluation and Theoretical Analysis for Representation Learning: A Surveycited by 5
  12. 2023A Universal Unbiased Method for Classification from Aggregate Observationscited by 5
24 more
  1. 2023Unified Risk Analysis for Weakly Supervised Learningcited by 5
  2. 2021Multi-Class Classification from Single-Class Data with Confidencescited by 3
  3. 2026Contrastive Identification and Generation in the Limitcited by 3
  4. 2021Labels, Information, and Computation: Efficient Learning Using Sufficient Labelscited by 2
  5. 2022Learning from Label Proportions with Instance-wise Consistencycited by 2
  6. 2023A Generalized Unbiased Risk Estimator for Learning with Augmented Classescited by 2
  7. 2023Feature Normalization Prevents Collapse of Non-contrastive Learning Dynamicscited by 2
  8. 2022Learning from Positive and Unlabeled Data with Augmented Classescited by 1
  9. 2024Robust Noisy Label Learning via Two-Stream Sample Distillationcited by 1
  10. 2025Unsupervised Pairwise Learning Optimization Framework for Cross-Corpus EEG-Based Emotion Recognition Based on Prototype Representationcited by 1
  11. 2021A Symmetric Loss Perspective of Reliable Machine Learningcited by 0
  12. 2023RoLNiP: Robust Learning Using Noisy Pairwise Comparisonscited by 0
  13. 2023Q&A Label Learningcited by 0
  14. 2024Learning from Concealed Labelscited by 0
  15. 2025Learning from M-Tuple Dominant Positive and Unlabeled Datacited by 0
  16. 2025A Unified Empirical Risk Minimization Framework for Flexible N-Tuples Weak Supervisioncited by 0
  17. 2025Learning from Similarity-Confidence and Confidence-Differencecited by 0
  18. 2025Learning from Uncertain Similarity and Unlabeled Datacited by 0
  19. 2025Learning from N-Tuple Data with M Positive Instances: Unbiased Risk Estimation and Theoretical Guaranteescited by 0
  20. 2025A Unified and Stable Risk Minimization Framework for Weakly Supervised Learning with Theoretical Guaranteescited by 0
  21. 2026FastBUS: A Fast Bayesian Framework for Unified Weakly-Supervised Learningcited by 0
  22. 2026Mixture Proportion Estimation and Weakly-supervised Kernel Test for Conditional Independencecited by 0
  23. 2026Embracing Biased Transition Matrices for Complementary-Label Learning with Many Classescited by 0
  24. 2026Importance Weighting for Unlabeled-unlabeled Learning under Distribution Shiftcited by 0