How to Allocate your Label Budget? Choosing between Active Learning and Learning to Reject in Anomaly Detection

TL;DR

This paper proposes a mixed strategy for allocating a limited label budget between active learning and learning to reject in anomaly detection, optimizing the use of labels to improve detection performance and user trust.

Contribution

It introduces a novel approach that dynamically allocates label budget between AL and LR based on a reward function, enhancing anomaly detection under limited labels.

Findings

Outperforms baseline strategies on 18 benchmark datasets

Effectively balances label allocation to improve detection accuracy

Increases user trust by optimizing rejection thresholds

Abstract

Anomaly detection attempts at finding examples that deviate from the expected behaviour. Usually, anomaly detection is tackled from an unsupervised perspective because anomalous labels are rare and difficult to acquire. However, the lack of labels makes the anomaly detector have high uncertainty in some regions, which usually results in poor predictive performance or low user trust in the predictions. One can reduce such uncertainty by collecting specific labels using Active Learning (AL), which targets examples close to the detector's decision boundary. Alternatively, one can increase the user trust by allowing the detector to abstain from making highly uncertain predictions, which is called Learning to Reject (LR). One way to do this is by thresholding the detector's uncertainty based on where its performance is low, which requires labels to be evaluated. Although both AL and LR need labels, they work with different types of labels: AL seeks strategic labels, which are evidently biased, while LR requires i.i.d. labels to evaluate the detector's performance and set the rejection threshold. Because one usually has a unique label budget, deciding how to optimally allocate it is challenging. In this paper, we propose a mixed strategy that, given a budget of labels, decides in multiple rounds whether to use the budget to collect AL labels or LR labels. The strategy is based on a reward function that measures the expected gain when allocating the budget to either side. We evaluate our strategy on 18 benchmark datasets and compare it to some baselines.