mlsanityData to 14 Oct 2026

Overcoming Overconfidence for Active Learning

arXiv 2308.10571 ↗First posted 21 Aug 2023cs.LG cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

It is not an exaggeration to say that the recent progress in artificial intelligence technology depends on large-scale and high-quality data. Simultaneously, a prevalent issue exists everywhere: the budget for data labeling is constrained. Active learning is a prominent approach for addressing this issue, where valuable data for labeling is selected through a model and utilized to iteratively adjust the model. However, due to the limited amount of data in each iteration, the model is vulnerable to bias; thus, it is more likely to yield overconfident predictions. In this paper, we present two novel methods to address the problem of overconfidence that arises in the active learning scenario. The first is an augmentation strategy named Cross-Mix-and-Mix (CMaM), which aims to calibrate the model by expanding the limited training distribution. The second is a selection strategy named Ranked Margin Sampling (RankedMS), which prevents choosing data that leads to overly confident predictions. Through various experiments and analyses, we are able to demonstrate that our proposals facilitate efficient data selection by alleviating overconfidence, even though they are readily applicable.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 23 of its 45 references that are in the map. First posted on arXiv 21 Aug 2023.

Builds on

23 of 45 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  4. 2017On Calibration of Modern Neural Networkscited by 4,960
  5. 2019When Does Label Smoothing Help?cited by 988
  6. 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 610
  7. 2019Learning Loss for Active Learningcited by 336
  8. 2019On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networkscited by 305
  9. 2019Variational Adversarial Active Learningcited by 284
  10. 2021Improving Calibration for Long-Tailed Recognitioncited by 212
  11. 2017Cost-Effective Active Learning for Deep Image Classificationcited by 166
  12. 2020Contextual Diversity for Active Learningcited by 139
11 more
  1. 2019Consistency-based Semi-supervised Active Learning: Towards Minimizing Labeling Costcited by 114
  2. 2020Task-Aware Variational Adversarial Active Learningcited by 84
  3. 2020Sequential Graph Convolutional Network for Active Learningcited by 69
  4. 2022Active Learning by Feature Mixingcited by 63
  5. 2020State-Relabeling Adversarial Active Learningcited by 60
  6. 2021Towards Calibrated Model for Long-Tailed Visual Recognition from Prior Perspectivecited by 39
  7. 2020VaB-AL: Incorporating Class Imbalance and Difficulty with Variational Bayes for Active Learningcited by 25
  8. 2022The AI Index 2022 Annual Reportcited by 19
  9. 2021MCDAL: Maximum Classifier Discrepancy for Active Learningcited by 18
  10. 2019SEAL: Semi-supervised Adversarial Active Learning on Attributed Graphscited by 11
  11. 2021Mitigating Sampling Bias and Improving Robustness in Active Learningcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 0; 2026: 2.

3 papers.

  1. 2024Materials-Discovery Workflows Guided by Symbolic Regression: Identifying Acid-Stable Oxides for Electrocatalysiscited by 0
  2. 2026A Critical Examination of Active Learning Workflows in Materials Sciencecited by 0
  3. 2026Learning Woody Clearing With Loss Alignment for Zero-Shot Regrowth and Woody Segmentationcited by 0