mlsanityData to 14 Oct 2026

Balancing Uncertainty and Diversity of Samples: Leveraging Diversity of Least, High Confidence Samples for Effective Active Learning

arXiv 2605.22169 ↗First posted 21 May 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep learning models, including Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), have achieved state-of-the-art performance on various computer vision tasks such as object classification, detection, segmentation, generation, and many more. However, these models are data-hungry as they require more training data to learn millions or billions of parameters. Especially for supervised learning tasks, curating a large number of labeled samples for model training is an expensive and time-consuming task. Active Learning (AL) has been used to address this problem for many years. Existing active learning methods aim at choosing the samples for annotation from a pool of unlabeled samples that are either diverse or uncertain. Choosing such samples may hinder the model's performance as we pool based on one dimension, i.e., either diverse or uncertain. In this paper, we propose four novel hybrid sampling methods for pooling both easy and hard samples, which are also diverse. To verify the efficacy of the proposed methods, extensive experiments are conducted using high and low-confidence samples separately. We observe from our experiments that the proposed hybrid sampling method, Least Confident and Diverse (LCD), consistently performs better compared to state-of-the-art methods. It is observed that selecting uncertain and diverse instances helps the model learn more distinct features. The codes related to this study will be available at https://github.com/XXX/LCD.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 43 of its 82 references that are in the map. First posted on arXiv 21 May 2026.

Builds on

43 of 82 references.

References · 43

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  5. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  6. 2016Densely Connected Convolutional Networkscited by 6,720
  7. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,859
  8. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  9. 2017Deep Bayesian Active Learning with Image Datacited by 819
  10. 2019Selection via Proxy: Efficient Data Selection for Deep Learningcited by 339
  11. 2019Variational Adversarial Active Learningcited by 284
  12. 2021Escaping the Big Data Paradigm with Compact Transformerscited by 264
31 more
  1. 2020Contextual Diversity for Active Learningcited by 139
  2. 2019Consistency-based Semi-supervised Active Learning: Towards Minimizing Labeling Costcited by 114
  3. 2021Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understandingcited by 97
  4. 2019ViewAL: Active Learning with Viewpoint Entropy for Semantic Segmentationcited by 96
  5. 2019Deep Active Learning: Unified and Principled Method for Query and Trainingcited by 93
  6. 2021Influence Selection for Active Learningcited by 84
  7. 2021Active Learning for Deep Object Detection via Probabilistic Modelingcited by 79
  8. 2017Generative Adversarial Active Learningcited by 70
  9. 2020Sequential Graph Convolutional Network for Active Learningcited by 69
  10. 2022Entropy-based Active Learning for Object Detection with Progressive Diversity Constraintcited by 53
  11. 2021Semi-Supervised Active Learning with Temporal Output Discrepancycited by 50
  12. 2017FSSD: Feature Fusion Single Shot Multibox Detectorcited by 43
  13. 2021Reducing Label Effort: Self-Supervised meets Active Learningcited by 42
  14. 2021Effective Evaluation of Deep Active Learning on Image Classification Taskscited by 37
  15. 2020Deep Active Learning for Biased Datasets via Fisher Kernel Self-Supervisioncited by 32
  16. 2019Semi-Supervised Learning with Scarce Annotationscited by 25
  17. 2020DEAL: Deep Evidential Active Learning for Image Classificationcited by 22
  18. 2020Training Deep Spiking Neural Networkscited by 21
  19. 2021Convergence of Uncertainty Sampling for Active Learningcited by 19
  20. 2019Combining MixMatch and Active Learning for Better Accuracy with Fewer Labelscited by 17
  21. 2021"BNN - BN = ?": Training Binary Neural Networks without Batch Normalizationcited by 17
  22. 2021Class-Balanced Active Learning for Image Classificationcited by 15
  23. 2017Feature-Fused SSD: Fast Detection for Small Objectscited by 14
  24. 2024Dataset Quantization with Active Learning based Adaptive Samplingcited by 13
  25. 2020Identifying Wrongly Predicted Samples: A Method for Active Learningcited by 8
  26. 2021Efficacy of Bayesian Neural Networks in Active Learningcited by 6
  27. 2023Army of Thieves: Enhancing Black-Box Model Extraction via Ensemble based sample selectioncited by 6
  28. 2022Zebra: Memory Bandwidth Reduction for CNN Accelerators With Zero Block Regularization of Activation Mapscited by 3
  29. 2019Dynamic Neural Network Channel Execution for Efficient Trainingcited by 2
  30. 2024Non-Uniform Illumination Attack for Fooling Convolutional Neural Networkscited by 2
  31. 2021Turning old models fashion again: Recycling classical CNN networks using the Lattice Transformationcited by 1

Built on by

Not cited yet.