mlsanityData to 14 Oct 2026

Rethinking Epistemic and Aleatoric Uncertainty for Active Open-Set Annotation: An Energy-Based Approach

arXiv 2502.19691 ↗First posted 27 Feb 2025cs.CV cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Active learning (AL), which iteratively queries the most informative examples from a large pool of unlabeled candidates for model training, faces significant challenges in the presence of open-set classes. Existing methods either prioritize query examples likely to belong to known classes, indicating low epistemic uncertainty (EU), or focus on querying those with highly uncertain predictions, reflecting high aleatoric uncertainty (AU). However, they both yield suboptimal performance, as low EU corresponds to limited useful information, and closed-set AU metrics for unknown class examples are less meaningful. In this paper, we propose an Energy-based Active Open-set Annotation (EAOA) framework, which effectively integrates EU and AU to achieve superior performance. EAOA features a (C+1)-class detector and a target classifier, incorporating an energy-based EU measure and a margin-based energy loss designed for the detector, alongside an energy-based AU measure for the target classifier. Another crucial component is the target-driven adaptive sampling strategy. It first forms a smaller candidate set with low EU scores to ensure closed-set properties, making AU metrics meaningful. Subsequently, examples with high AU scores are queried to form the final query set, with the candidate set size adjusted adaptively. Extensive experiments show that EAOA achieves state-of-the-art performance while maintaining high query precision and low training overhead. The code is available at https://github.com/chenchenzong/EAOA.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 29 of its 58 references that are in the map. First posted on arXiv 27 Feb 2025.

Builds on

29 of 58 references.

References · 29

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  3. 2020Energy-based Out-of-distribution Detectioncited by 1,299
  4. 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 610
  5. 2020A Survey of Deep Active Learningcited by 554
  6. 2021Open-Set Recognition: a Good Closed-Set Classifier is All You Need?cited by 353
  7. 2019Learning Loss for Active Learningcited by 336
  8. 2014Towards Open World Recognitioncited by 226
  9. 2021Deep Deterministic Uncertainty: A Simple Baselinecited by 200
  10. 2018Understanding Measures of Uncertainty for Adversarial Example Detectioncited by 177
  11. 2017Generative OpenMax for Multi-Class Open Set Classificationcited by 157
  12. 2021Adversarial Reciprocal Points Learning for Open Set Recognitioncited by 153
17 more
  1. 2019C2AE: Class Conditioned Auto-Encoder for Open-set Recognitioncited by 142
  2. 2021Learning Placeholders for Open-Set Recognitioncited by 121
  3. 2020Learning Open Set Network with Discriminative Reciprocal Pointscited by 115
  4. 2020Hybrid Models for Open Set Recognitioncited by 109
  5. 2020Conditional Gaussian Distribution Learning for Open Set Recognitioncited by 104
  6. 2021Can multi-label classification networks know what they don't know?cited by 100
  7. 2017Learning how to Active Learn: A Deep Reinforcement Learning Approachcited by 79
  8. 2018Efficient Active Learning for Image Classification and Segmentation using a Sample Selection and Conditional Generative Adversarial Networkcited by 74
  9. 2017Generative Adversarial Active Learningcited by 70
  10. 2018Adversarial Sampling for Active Learningcited by 51
  11. 2017Sparse Representation-based Open Set Recognitioncited by 36
  12. 2023Active Finetuning: Exploiting Annotation Budget in the Pretraining-Finetuning Paradigmcited by 36
  13. 2022Active Learning for Open-set Annotationcited by 34
  14. 2022Meta-Query-Net: Resolving Purity-Informativeness Dilemma in Open-set Active Learningcited by 25
  15. 2023Entropic Open-set Active Learningcited by 23
  16. 2024Dirichlet-Based Prediction Calibration for Learning with Noisy Labelscited by 8
  17. 2024Bidirectional Uncertainty-Based Active Learning for Open Set Annotationcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 4.

5 papers.

  1. 2026Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learningcited by 1
  2. 2026Delving Aleatoric Uncertainty in Medical Image Segmentation via Vision Foundation Modelscited by 1
  3. 2025Combining Discrepancy-Confusion Uncertainty and Calibration Diversity for Active Fine-Grained Image Classificationcited by 0
  4. 2026Softmax is not Enough (for Adaptive Conformal Classification)cited by 0
  5. 2026Federated Active Learning Under Extreme Non-IID and Global Class Imbalancecited by 0