# A Variance Maximization Criterion for Active Learning

arXiv: https://arxiv.org/abs/1706.07642 · HTML: https://mlsanity.com/paper/1706.07642/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jun 2017.

## Built on by (most cited first, 9 of 9)

- [Reducing Label Effort: Self-Supervised meets Active Learning](https://mlsanity.com/paper/2108.11458/) (2021)
- [PT4AL: Using Self-Supervised Pretext Tasks for Active Learning](https://mlsanity.com/paper/2201.07459/) (2022)
- [Class-Balanced Active Learning for Image Classification](https://mlsanity.com/paper/2110.04543/) (2021)
- [When Deep Learners Change Their Mind: Learning Dynamics for Active Learning](https://mlsanity.com/paper/2107.14707/) (2021)
- [Uncertainty Quantification in Multivariable Regression for Material Property Prediction with Bayesian Neural Networks](https://mlsanity.com/paper/2311.02495/) (2023)
- [A Unified Active Learning Framework for Annotating Graph Data with Application to Software Source Code Performance Prediction](https://mlsanity.com/paper/2304.13032/) (2023)
- [Addressing practical challenges in Active Learning via a hybrid query strategy](https://mlsanity.com/paper/2110.03785/) (2021)
- [Dynamic Exploration-Exploitation Trade-Off in Active Learning Regression with Bayesian Hierarchical Modeling](https://mlsanity.com/paper/2304.07665/) (2023)
- [Graceful Degradation and Related Fields](https://mlsanity.com/paper/2106.11119/) (2021)

## Abstract (arXiv)

Active learning aims to train a classifier as fast as possible with as few labels as possible. The core element in virtually any active learning strategy is the criterion that measures the usefulness of the unlabeled data based on which new points to be labeled are picked. We propose a novel approach which we refer to as maximizing variance for active learning or MVAL for short. MVAL measures the value of unlabeled instances by evaluating the rate of change of output variables caused by changes in the next sample to be queried and its potential labelling. In a sense, this criterion measures how unstable the classifier's output is for the unlabeled data points under perturbations of the training data. MVAL maintains, what we refer to as, retraining information matrices to keep track of these output scores and exploits two kinds of variance to measure the informativeness and representativeness, respectively. By fusing these variances, MVAL is able to select the instances which are both informative and representative. We employ our technique both in combination with logistic regression and support vector machines and demonstrate that MVAL achieves state-of-the-art performance in experiments on a large number of standard benchmark datasets.