mlsanityData to 7 Oct 2026

Deep Active Learning with Noise Stability

arXiv 2205.13340 ↗First posted 26 May 2022cs.LGcited by 13

Not on the map. Why?

Abstract (arXiv)

Uncertainty estimation for unlabeled data is crucial to active learning. With a deep neural network employed as the backbone model, the data selection process is highly challenging due to the potential over-confidence of the model inference. Existing methods resort to special learning fashions (e.g. adversarial) or auxiliary models to address this challenge. This tends to result in complex and inefficient pipelines, which would render the methods impractical. In this work, we propose a novel algorithm that leverages noise stability to estimate data uncertainty. The key idea is to measure the output derivation from the original observation when the model parameters are randomly perturbed by noise. We provide theoretical analyses by leveraging the small Gaussian noise theory and demonstrate that our method favors a subset with large and diverse gradients. Our method is generally applicable in various tasks, including computer vision, natural language processing, and structural data analysis. It achieves competitive performance compared against state-of-the-art active learning baselines.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 35 of its 78 references that are in the map. First posted on arXiv 26 May 2022.

Builds on

35 of 78 references.

References · 35

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  5. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  7. 2017Deep Bayesian Active Learning with Image Datacited by 812
  8. 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 602
  9. 2020A Survey of Deep Active Learningcited by 547
  10. 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 428
  11. 2019Learning Loss for Active Learningcited by 335
  12. 2018Stronger generalization bounds for deep nets via a compression approachcited by 306
23 more
  1. 2017Dilated Residual Networkscited by 293
  2. 2019Variational Adversarial Active Learningcited by 284
  3. 2018Adversarial Active Learning for Deep Networks: a Margin Based Approachcited by 155
  4. 2021Batch Active Learning at Scalecited by 127
  5. 2019A General Framework for Uncertainty Estimation in Deep Learningcited by 119
  6. 2021Scalable Marginal Likelihood Estimation for Model Selection in Deep Learningcited by 105
  7. 2017Learning Active Learning from Datacited by 100
  8. 2021Gone Fishing: Neural Active Learning with Fisher Embeddingscited by 88
  9. 2020Bayesian Deep Learning via Subnetwork Inferencecited by 85
  10. 2021Influence Selection for Active Learningcited by 82
  11. 2018Efficient Active Learning for Image Classification and Segmentation using a Sample Selection and Conditional Generative Adversarial Networkcited by 74
  12. 2017Generative Adversarial Active Learningcited by 69
  13. 2020Sequential Graph Convolutional Network for Active Learningcited by 69
  14. 2020State-Relabeling Adversarial Active Learningcited by 60
  15. 2018Adversarial Sampling for Active Learningcited by 51
  16. 2018CEREALS - Cost-Effective REgion-based Active Learning for Semantic Segmentationcited by 47
  17. 2017Cost-Effective Active Learning for Melanoma Segmentationcited by 40
  18. 2021Effective Evaluation of Deep Active Learning on Image Classification Taskscited by 37
  19. 2018Cost-Sensitive Active Learning for Intracranial Hemorrhage Detectioncited by 28
  20. 2016Active Learning for Speech Recognition: the Power of Gradientscited by 25
  21. 2015Hierarchical Subquery Evaluation for Active Learning on a Graphcited by 21
  22. 2021Boosting Active Learning via Improving Test Performancecited by 19
  23. 2020Ask-n-Learn: Active Learning via Reliable Gradient Representations for Image Classificationcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 3; 2025: 5; 2026: 3.

13 papers.

Sort:

By area:

  1. 2023You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentationcited by 6
  2. 2023Deep Active Learning in the Presence of Label Noise: A Surveycited by 5
  3. 2026Next Generation Active Learning: Mixture of LLMs in the Loopcited by 3
  4. 2024Proactive Schemes: A Survey of Adversarial Attacks for Social Goodcited by 2
  5. 2024LPLgrad: Optimizing Active Learning Through Gradient Norm Sample Selection and Auxiliary Model Trainingcited by 2
  6. 2026VIOLA: Towards Video In-Context Learning with Minimal Annotationscited by 2
  7. 2025ALScope: A Unified Toolkit for Deep Active Learningcited by 1
  8. 2024Salutary Labeling with Zero Human Annotationcited by 0
  9. 2025A Novel Active Learning Approach to Label One Million Unknown Malware Variantscited by 0
  10. 2025Combining Discrepancy-Confusion Uncertainty and Calibration Diversity for Active Fine-Grained Image Classificationcited by 0
  11. 2025Unsupervised Active Learning via Natural Feature Progressive Frameworkcited by 0
  12. 2025Uncertainty-Aware Subset Selection for Robust Visual Explainability under Distribution Shiftscited by 0
1 more
  1. 2026FACT: A Simple and Efficient Framework for Active Finetuningcited by 0