BoSS: A Best-of-Strategies Selector as an Oracle for Deep Active Learning

TL;DR

BoSS is a scalable oracle strategy for deep active learning that combines multiple selection strategies to identify the most beneficial data batches, outperforming existing methods and providing a robust benchmark for future research.

Contribution

Introduces BoSS, a scalable ensemble-based oracle strategy for deep active learning that improves selection robustness and performance across large datasets and models.

Findings

BoSS outperforms existing oracle strategies.

State-of-the-art AL strategies lag behind oracle performance.

Ensemble approaches can mitigate AL strategy inconsistencies.

Abstract

Active learning (AL) aims to reduce annotation costs while maximizing model performance by iteratively selecting valuable instances. While foundation models have made it easier to identify these instances, existing selection strategies still lack robustness across different models, annotation budgets, and datasets. To highlight the potential weaknesses of existing AL strategies and provide a reference point for research, we explore oracle strategies, i.e., strategies that approximate the optimal selection by accessing ground-truth information unavailable in practical AL scenarios. Current oracle strategies, however, fail to scale effectively to large datasets and complex deep neural networks. To tackle these limitations, we introduce the Best-of-Strategy Selector (BoSS), a scalable oracle strategy designed for large-scale AL scenarios. BoSS constructs a set of candidate batches through an ensemble of selection strategies and then selects the batch yielding the highest performance gain. As an ensemble of selection strategies, BoSS can be easily extended with new state-of-the-art strategies as they emerge, ensuring it remains a reliable oracle strategy in the future. Our evaluation demonstrates that i) BoSS outperforms existing oracle strategies, ii) state-of-the-art AL strategies still fall noticeably short of oracle performance, especially in large-scale datasets with many classes, and iii) one possible solution to counteract the inconsistent performance of AL strategies might be to employ an ensemble-based approach for the selection.