# When Deep Learners Change Their Mind: Learning Dynamics for Active Learning

arXiv: https://arxiv.org/abs/2107.14707 · HTML: https://mlsanity.com/paper/2107.14707/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 33 references that are in the map.
- First posted on arXiv 30 Jul 2021.

## Builds on (most cited first, 17 of 17)

- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Deep Bayesian Active Learning with Image Data](https://mlsanity.com/paper/1703.02910/) (2017)
- [An Empirical Study of Example Forgetting during Deep Neural Network Learning](https://mlsanity.com/paper/1812.05159/) (2018)
- [Learning Loss for Active Learning](https://mlsanity.com/paper/1905.03677/) (2019)
- [Variational Adversarial Active Learning](https://mlsanity.com/paper/1904.00370/) (2019)
- [Cost-Effective Active Learning for Deep Image Classification](https://mlsanity.com/paper/1701.03551/) (2017)
- [Efficient Active Learning for Image Classification and Segmentation using a Sample Selection and Conditional Generative Adversarial Network](https://mlsanity.com/paper/1806.05473/) (2018)
- [Generative Adversarial Active Learning](https://mlsanity.com/paper/1702.07956/) (2017)
- [Active Learning for Deep Detection Neural Networks](https://mlsanity.com/paper/1911.09168/) (2019)
- [Adversarial Sampling for Active Learning](https://mlsanity.com/paper/1808.06671/) (2018)
- [Active Learning for Deep Object Detection](https://mlsanity.com/paper/1809.09875/) (2018)
- [Temporal Coherence for Active Learning in Videos](https://mlsanity.com/paper/1908.11757/) (2019)
- [Large-Scale Visual Active Learning with Deep Probabilistic Ensembles](https://mlsanity.com/paper/1811.03575/) (2018)
- [Active Multi-Kernel Domain Adaptation for Hyperspectral Image Classification](https://mlsanity.com/paper/1904.05200/) (2019)
- [Learning to Rank for Active Learning: A Listwise Approach](https://mlsanity.com/paper/2008.00078/) (2020)
- [A Variance Maximization Criterion for Active Learning](https://mlsanity.com/paper/1706.07642/) (2017)
- [Active Transfer Learning with Zero-Shot Priors: Reusing Past Datasets for Future Tasks](https://mlsanity.com/paper/1510.01544/) (2015)

## Built on by (most cited first, 6 of 6)

- [Reducing Label Effort: Self-Supervised meets Active Learning](https://mlsanity.com/paper/2108.11458/) (2021)
- [Warm Start Active Learning with Proxy Labels & Selection via Semi-Supervised Fine-Tuning](https://mlsanity.com/paper/2209.06285/) (2022)
- [Class-Balanced Active Learning for Image Classification](https://mlsanity.com/paper/2110.04543/) (2021)
- [Radically Lower Data-Labeling Costs for Visually Rich Document Extraction Models](https://mlsanity.com/paper/2210.16391/) (2022)
- [CQural: A Novel CNN based Hybrid Architecture for Quantum Continual Machine Learning](https://mlsanity.com/paper/2305.09738/) (2023)
- [Advancing Vision-based Human Action Recognition: Exploring Vision-Language CLIP Model for Generalisation in Domain-Independent Tasks](https://mlsanity.com/paper/2507.18675/) (2025)

## Abstract (arXiv)

Active learning aims to select samples to be annotated that yield the largest performance improvement for the learning algorithm. Many methods approach this problem by measuring the informativeness of samples and do this based on the certainty of the network predictions for samples. However, it is well-known that neural networks are overly confident about their prediction and are therefore an untrustworthy source to assess sample informativeness. In this paper, we propose a new informativeness-based active learning method. Our measure is derived from the learning dynamics of a neural network. More precisely we track the label assignment of the unlabeled data pool during the training of the algorithm. We capture the learning dynamics with a metric called label-dispersion, which is low when the network consistently assigns the same label to the sample during the training of the network and high when the assigned label changes frequently. We show that label-dispersion is a promising predictor of the uncertainty of the network, and show on two benchmark datasets that an active learning algorithm based on label-dispersion obtains excellent results.