Adaptive Active Learning for Regression via Reinforcement Learning
This work addresses labeling efficiency for regression tasks, particularly in domains with irregular data density, but it is incremental as it builds on an existing method.
The paper tackled the problem of reducing labeling costs in active learning for regression by proposing Weighted improved Greedy Sampling (WiGS), which uses reinforcement learning to dynamically adjust sample selection, and it outperformed baseline methods on 18 benchmark datasets with improved accuracy and efficiency.
Active learning for regression reduces labeling costs by selecting the most informative samples. Improved Greedy Sampling is a prominent method that balances feature-space diversity and output-space uncertainty using a static, multiplicative rule. We propose Weighted improved Greedy Sampling (WiGS), which replaces this framework with a dynamic, additive criterion. We formulate weight selection as a reinforcement learning problem, enabling an agent to adapt the exploration-investigation balance throughout learning. Experiments on 18 benchmark datasets and a synthetic environment show WiGS outperforms iGS and other baseline methods in both accuracy and labeling efficiency, particularly in domains with irregular data density where the baseline's multiplicative rule ignores high-error samples in dense regions.