Active Nearest-Neighbor Learning in Metric Spaces
This work addresses label efficiency in non-parametric classification for machine learning practitioners, offering a novel active learning approach.
The paper tackles the problem of active learning in metric spaces by proposing MARMANN, an algorithm that achieves competitive prediction error guarantees with significantly lower label complexity than passive learners.
We propose a pool-based non-parametric active learning algorithm for general metric spaces, called MArgin Regularized Metric Active Nearest Neighbor (MARMANN), which outputs a nearest-neighbor classifier. We give prediction error guarantees that depend on the noisy-margin properties of the input sample, and are competitive with those obtained by previously proposed passive learners. We prove that the label complexity of MARMANN is significantly lower than that of any passive learner with similar error guarantees. MARMANN is based on a generalized sample compression scheme, and a new label-efficient active model-selection procedure.