Pool-Based Sequential Active Learning for Regression

TL;DR

This paper introduces a new pool-based sequential active learning method for regression that emphasizes informativeness, representativeness, and diversity, and demonstrates its effectiveness through extensive experiments.

Contribution

It proposes a novel ALR approach combining passive sampling with existing methods, enhancing sample selection for regression tasks.

Findings

The new approach outperforms existing ALR methods on multiple datasets.

Incorporating representativeness and diversity improves sample efficiency.

The method is versatile and can be integrated with other ALR approaches.

Abstract

Active learning is a machine learning approach for reducing the data labeling effort. Given a pool of unlabeled samples, it tries to select the most useful ones to label so that a model built from them can achieve the best possible performance. This paper focuses on pool-based sequential active learning for regression (ALR). We first propose three essential criteria that an ALR approach should consider in selecting the most useful unlabeled samples: informativeness, representativeness, and diversity, and compare four existing ALR approaches against them. We then propose a new ALR approach using passive sampling, which considers both the representativeness and the diversity in both the initialization and subsequent iterations. Remarkably, this approach can also be integrated with other existing ALR approaches in the literature to further improve the performance. Extensive experiments on 11 UCI, CMU StatLib, and UFL Media Core datasets from various domains verified the effectiveness of our proposed ALR approaches.