mlsanityData to 7 Oct 2026

Influence Selection for Active Learning

arXiv 2108.09331 ↗First posted 20 Aug 2021cs.LG cs.CVcited by 82

Not on the map. Why?

Abstract (arXiv)

The existing active learning methods select the samples by evaluating the sample's uncertainty or its effect on the diversity of labeled datasets based on different task-specific or model-specific criteria. In this paper, we propose the Influence Selection for Active Learning(ISAL) which selects the unlabeled samples that can provide the most positive Influence on model performance. To obtain the Influence of the unlabeled sample in the active learning scenario, we design the Untrained Unlabeled sample Influence Calculation(UUIC) to estimate the unlabeled sample's expected gradient with which we calculate its Influence. To prove the effectiveness of UUIC, we provide both theoretical and experimental analyses. Since the UUIC just depends on the model gradients, which can be obtained easily from any neural network, our active learning algorithm is task-agnostic and model-agnostic. ISAL achieves state-of-the-art performance in different active learning settings for different tasks with different datasets. Compared with previous methods, our method decreases the annotation cost at least by 12%, 13% and 16% on CIFAR10, VOC2012 and COCO, respectively.

Key facts. Cited by 82 papers in the map (data through 7 Oct 2026). Builds on 19 of its 42 references that are in the map. First posted on arXiv 20 Aug 2021.

Builds on

19 of 42 references.

References · 19

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  4. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  5. 2015SSD: Single Shot MultiBox Detectorcited by 3,125
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  7. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,573
  8. 2019MMDetection: Open MMLab Detection Toolbox and Benchmarkcited by 1,379
  9. 2020A Survey of Deep Active Learningcited by 547
  10. 2019Learning Loss for Active Learningcited by 335
  11. 2018Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problemcited by 323
  12. 2019Variational Adversarial Active Learningcited by 284
7 more
  1. 2017Cost-Effective Active Learning for Deep Image Classificationcited by 165
  2. 2019On Network Design Spaces for Visual Recognitioncited by 68
  3. 2020Scalable Active Learning for Object Detectioncited by 58
  4. 2020Not All Unlabeled Data are Equal: Learning to Weight Data in Semi-supervised Learningcited by 53
  5. 2018Localization-Aware Active Learning for Object Detectioncited by 48
  6. 2015Hierarchical Subquery Evaluation for Active Learning on a Graphcited by 21
  7. 2016Second-Order Stochastic Optimization for Machine Learning in Linear Timecited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 11; 2023: 20; 2024: 21; 2025: 18; 2026: 9.

82 papers.

Sort:

By area:

  1. 2023TRAK: Attributing Model Behavior at Scalecited by 273
  2. 2022Training Data Influence Analysis and Estimation: A Surveycited by 134
  3. 2024What is Your Data Worth to GPT? LLM-Scale Data Valuation with Influence Functionscited by 88
  4. 2023Intriguing Properties of Data Attribution on Diffusion Modelscited by 47
  5. 2023Active Finetuning: Exploiting Annotation Budget in the Pretraining-Finetuning Paradigmcited by 36
  6. 2024Training Data Attribution via Approximate Unrolled Differentiationcited by 33
  7. 2021Deep Active Learning by Leveraging Training Dynamicscited by 32
  8. 2023A comprehensive survey on deep active learning in medical image analysiscited by 32
  9. 2022PT4AL: Using Self-Supervised Pretext Tasks for Active Learningcited by 27
  10. 2023Exploring Active 3D Object Detection from a Generalization Perspectivecited by 26
  11. 2024Diffusion Attribution Score: Evaluating Training Data Influence in Diffusion Modelscited by 21
  12. 2021Boosting Active Learning via Improving Test Performancecited by 19
38 more
  1. 2022Active Learning Strategies for Weakly-supervised Object Detectioncited by 17
  2. 2022Semantic Segmentation with Active Semi-Supervised Learningcited by 16
  3. 2022Adapting and Evaluating Influence-Estimation Methods for Gradient-Boosted Decision Treescited by 16
  4. 2024ActiveAD: Planning-Oriented Active Learning for End-to-End Autonomous Drivingcited by 16
  5. 2022Active Pointly-Supervised Instance Segmentationcited by 15
  6. 2022Deep Active Learning with Noise Stabilitycited by 13
  7. 2023KECOR: Kernel Coding Rate Maximization for Active 3D Object Detectioncited by 13
  8. 2023Divide and Adapt: Active Domain Adaptation via Customized Learningcited by 13
  9. 2024Data Attribution for Diffusion Models: Timestep-induced Bias in Influence Estimationcited by 13
  10. 2024Fairness Without Harm: An Influence-Guided Active Sampling Approachcited by 12
  11. 2023Towards Free Data Selection with General-Purpose Modelscited by 10
  12. 2024Neural Active Learning Beyond Banditscited by 10
  13. 2024Outlier Gradient Analysis: Efficiently Identifying Detrimental Training Samples for Deep Learning Modelscited by 10
  14. 2024Generative Active Learning for Long-tailed Instance Segmentationcited by 10
  15. 2023Towards Balanced Active Learning for Multimodal Classificationcited by 9
  16. 2023BAL: Balancing Diversity and Novelty for Active Learningcited by 9
  17. 2023Error Discovery by Clustering Influence Embeddingscited by 7
  18. 2025Uncertainty Meets Diversity: A Comprehensive Active Learning Framework for Indoor 3D Object Detectioncited by 7
  19. 2021Towards General and Efficient Active Learningcited by 6
  20. 2024Sparse, Efficient and Explainable Data Attribution with DualXDAcited by 6
  21. 2023A Simple and Efficient Baseline for Data Attribution on Imagescited by 5
  22. 2024Improve Knowledge Distillation via Label Revision and Data Selectioncited by 5
  23. 2025First is Not Really Better Than Last: Evaluating Layer Choice and Aggregation Strategies in Language Model Data Influence Estimationcited by 5
  24. 2023Leveraging Importance Weights in Subset Selectioncited by 4
  25. 2025Nonparametric Data Attribution for Diffusion Modelscited by 4
  26. 2022Bucketized Active Sampling for Learning ACOPFcited by 3
  27. 2023Active Learning with Combinatorial Coveragecited by 3
  28. 2023Automated Detection of Cat Facial Landmarkscited by 3
  29. 2024Composite Active Learning: Towards Multi-Domain Active Learning with Theoretical Guaranteescited by 3
  30. 2025Unraveling Indirect In-Context Learning Using Influence Functionscited by 3
  31. 2025Distributional Training Data Attribution: What do Influence Functions Sample?cited by 3
  32. 2025Learn 3D VQA Better with Active Selection and Reannotationcited by 3
  33. 2025Understanding Data Influence with Differential Approximationcited by 3
  34. 2022Pixel is All You Need: Adversarial Trajectory-Ensemble Active Learning for Salient Object Detectioncited by 2
  35. 2023Transferable Candidate Proposal with Bounded Uncertaintycited by 2
  36. 2024VeCAF: Vision-language Collaborative Active Finetuning with Training Objective Awarenesscited by 2
  37. 2024Direct Acquisition Optimization for Low-Budget Active Learningcited by 2
  38. 2024Enhancing Semi-Supervised Learning via Representative and Diverse Sample Selectioncited by 2

50 most cited of 82 All 82 (JSON)