A Variance Maximization Criterion for Active Learning

TL;DR

This paper introduces MVAL, a novel active learning criterion that evaluates the potential informativeness and representativeness of unlabeled data by measuring the instability of classifier outputs, leading to improved data selection.

Contribution

The paper proposes MVAL, a new variance-based criterion for active learning that combines informativeness and representativeness to select optimal unlabeled instances.

Findings

MVAL achieves state-of-the-art performance on benchmark datasets.

It effectively balances informativeness and representativeness.

Demonstrates improvements with logistic regression and SVMs.

Abstract

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.

Full text

See pages 1-last of MVAL.pdf