mlsanityData to 7 Oct 2026

Not All Unlabeled Data are Equal: Learning to Weight Data in Semi-supervised Learning

arXiv 2007.01293 ↗First posted 2 Jul 2020cs.LG cs.CV stat.MLcited by 53

Not on the map. Why?

Abstract (arXiv)

Existing semi-supervised learning (SSL) algorithms use a single weight to balance the loss of labeled and unlabeled examples, i.e., all unlabeled examples are equally weighted. But not all unlabeled data are equal. In this paper we study how to use a different weight for every unlabeled example. Manual tuning of all those weights -- as done in prior work -- is no longer possible. Instead, we adjust those weights via an algorithm based on the influence function, a measure of a model's dependency on one training example. To make the approach efficient, we propose a fast and effective approximation of the influence function. We demonstrate that this technique outperforms state-of-the-art methods on semi-supervised image and language classification tasks.

Key facts. Cited by 53 papers in the map (data through 7 Oct 2026). First posted on arXiv 2 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 12; 2022: 22; 2023: 9; 2024: 6; 2025: 3; 2026: 1.

53 papers.

Sort:

By area:

  1. 2021A Survey on Deep Semi-supervised Learningcited by 250
  2. 2021A Survey of Human-in-the-loop for Machine Learningcited by 214
  3. 2021ST++: Make Self-training Work Better for Semi-supervised Semantic Segmentationcited by 184
  4. 2021Dash: Semi-Supervised Learning with Dynamic Thresholdingcited by 147
  5. 2023SoftMatch: Addressing the Quantity-Quality Trade-off in Semi-supervised Learningcited by 133
  6. 2021Influence Selection for Active Learningcited by 82
  7. 2022Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentationcited by 74
  8. 2021DASO: Distribution-Aware Semantics-Oriented Pseudo-label for Imbalanced Semi-Supervised Learningcited by 51
  9. 2022Unbiased Subclass Regularization for Semi-Supervised Semantic Segmentationcited by 33
  10. 2022Betty: An Automatic Differentiation Library for Multilevel Optimizationcited by 27
  11. 2022In Defense of Core-set: A Density-aware Core-set Selection for Active Learningcited by 26
  12. 2022Equivariance Discovery by Learned Parameter-Sharingcited by 18
38 more
  1. 2023IMMA: Immunizing text-to-image Models against Malicious Adaptationcited by 17
  2. 2022FairIF: Boosting Fairness in Deep Learning via Influence Functions with Validation Set Sensitive Attributescited by 12
  3. 2022An Embarrassingly Simple Baseline for Imbalanced Semi-Supervised Learningcited by 12
  4. 2022Fix-A-Step: Semi-supervised Learning from Uncurated Unlabeled Datacited by 11
  5. 2022Label-efficient Hybrid-supervised Learning for Medical Image Segmentationcited by 10
  6. 2023RoPAWS: Robust Semi-supervised Representation Learning from Uncurated Datacited by 10
  7. 2025Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safetycited by 10
  8. 2021CA-SSL: Class-Agnostic Semi-Supervised Learning for Detection and Segmentationcited by 9
  9. 2022MaxMatch: Semi-Supervised Learning with Worst-Case Consistencycited by 9
  10. 2022Total Variation Optimization Layers for Computer Visioncited by 8
  11. 2022Conformal Credal Self-Supervised Learningcited by 8
  12. 2022Learning to Predict Gradients for Semi-Supervised Continual Learningcited by 7
  13. 2023Out-of-Distributed Semantic Pruning for Robust Semi-Supervised Learningcited by 7
  14. 2022On the Importance of Calibration in Semi-supervised Learningcited by 6
  15. 2021FLiText: A Faster and Lighter Semi-Supervised Text Classification with Convolution Networkscited by 5
  16. 2022Semi-supervised Learning using Robust Losscited by 4
  17. 2022CEIP: Combining Explicit and Implicit Priors for Reinforcement Learning with Demonstrationscited by 4
  18. 2023Local or Global: Selective Knowledge Assimilation for Federated Learning with Limited Labelscited by 4
  19. 2022Robust Deep Semi-Supervised Learning: A Brief Introductioncited by 3
  20. 2022Ask2Mask: Guided Data Selection for Masked Speech Modelingcited by 3
  21. 2022Open-Set Semi-Supervised Learning for 3D Point Cloud Understandingcited by 3
  22. 2022Confident Sinkhorn Allocation for Pseudo-Labelingcited by 3
  23. 2023Unlabeled Imperfect Demonstrations in Adversarial Imitation Learningcited by 3
  24. 2025OmViD: Omni-supervised active learning for video action detectioncited by 3
  25. 2024Generalizable and Stable Finetuning of Pretrained Language Models on Low-Resource Textscited by 2
  26. 2024Forming Auxiliary High-confident Instance-level Loss to Promote Learning from Label Proportionscited by 2
  27. 2021Automated Feature-Specific Tree Species Identification from Natural Images using Deep Semi-Supervised Learningcited by 1
  28. 2021Semi-supervised teacher-student deep neural network for materials discoverycited by 1
  29. 2023Pruning the Unlabeled Data to Improve Semi-Supervised Learningcited by 1
  30. 2023Towards Self-Adaptive Pseudo-Label Filtering for Semi-Supervised Learningcited by 1
  31. 2021Learning from Mistakes based on Class Weighting with Application to Neural Architecture Searchcited by 0
  32. 2021Taming Overconfident Prediction on Unlabeled Data from Hindsightcited by 0
  33. 2022Few Clean Instances Help Denoising Distant Supervisioncited by 0
  34. 2022End to End Generative Meta Curriculum Learning For Medical Data Augmentationcited by 0
  35. 2023On Training Implicit Meta-Learning With Applications to Inductive Weighing in Consistency Regularizationcited by 0
  36. 2024Uncertainty-aware Sampling for Long-tailed Semi-supervised Learningcited by 0
  37. 2024Diverse Teacher-Students for Deep Safe Semi-Supervised Learning under Class Mismatchcited by 0
  38. 2024LayerMatch: Do Pseudo-labels Benefit All Layers?cited by 0

50 most cited of 53 All 53 (JSON)