Gradient-Discrepancy Acquisition for Pool-Based Active Learning
For practitioners of active learning, this provides a theoretically grounded and effective criterion for selecting informative data points.
The paper proposes a novel gradient-based acquisition criterion for pool-based active learning, derived from a generalization bound, and shows it outperforms existing methods in empirical evaluations.
The effectiveness of active learning hinges on the choice of the acquisition criterion by which a learning algorithm selects potentially informative data points whose label is subsequently queried. This paper proposes a novel gradient-based acquisition criterion, derived from a generalization bound introduced by Luo et al. (2022). This criterion can be applied in lieu of uncertainty measures in uncertainty sampling, or incorporated into diversity-based methods that consider the spread of sampled points in addition to the uncertainty of their labels. We provide a theoretical justification of the proposed acquisition criterion, and demonstrate its effectiveness in an empirical evaluation.