PercentMatch: Percentile-based Dynamic Thresholding for Multi-Label Semi-Supervised Classification
Not on the map. Why?
Abstract (arXiv)
While much of recent study in semi-supervised learning (SSL) has achieved strong performance on single-label classification problems, an equally important yet underexplored problem is how to leverage the advantage of unlabeled data in multi-label classification tasks. To extend the success of SSL to multi-label classification, we first analyze with illustrative examples to get some intuition about the extra challenges exist in multi-label classification. Based on the analysis, we then propose PercentMatch, a percentile-based threshold adjusting scheme, to dynamically alter the score thresholds of positive and negative pseudo-labels for each class during the training, as well as dynamic unlabeled loss weights that further reduces noise from early-stage unlabeled predictions. Without loss of simplicity, we achieve strong performance on Pascal VOC2007 and MS-COCO datasets when compared to recent SSL methods.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 28 of its 38 references that are in the map. First posted on arXiv 30 Aug 2022.
Builds on
28 of 38 references.
References · 28
- 2014Microsoft COCO: Common Objects in Contextcited by 17,608
- 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,085
- 2019MixMatch: A Holistic Approach to Semi-Supervised Learningcited by 1,391
- 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,060
- 2016Temporal Ensembling for Semi-Supervised Learningcited by 1,057
- 2018Exploring the Limits of Weakly Supervised Pretrainingcited by 605
- 2017Deep Learning Scaling is Predictable, Empiricallycited by 582
- 2021FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labelingcited by 540
- 2019Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learningcited by 510
- 2017Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning resultscited by 472
16 more
- 2016Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learningcited by 441
- 2020Asymmetric Loss For Multi-Label Classificationcited by 381
- 2021In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learningcited by 301
- 2019On The Power of Curriculum Learning in Training Deep Networkscited by 286
- 2021A Survey on Deep Semi-supervised Learningcited by 251
- 2019Billion-scale semi-supervised learning for image classificationcited by 232
- 2022FreeMatch: Self-adaptive Thresholding for Semi-supervised Learningcited by 232
- 2020Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentationcited by 222
- 2022Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labelscited by 218
- 2021Dash: Semi-Supervised Learning with Dynamic Thresholdingcited by 148
- 2020When Do Curricula Work?cited by 106
- 2021ML-Decoder: Scalable and Versatile Classification Headcited by 79
- 2021Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detectioncited by 27
- 2019Distance-Based Learning from Errors for Confidence Calibrationcited by 23
- 2021Self-supervised Mean Teacher for Semi-supervised Chest X-ray Classificationcited by 14
- 2020Negative Pseudo Labeling using Class Proportion for Semantic Segmentation in Pathologycited by 12
Built on by
Citing papers per year: 2023: 1.
1 paper.