Nonparametric active learning for cost-sensitive classification

TL;DR

This paper introduces a nonparametric active learning algorithm tailored for cost-sensitive classification, achieving optimal convergence rates and explicitly quantifying gains over passive learning under certain noise conditions.

Contribution

It presents a novel active learning method for cost-sensitive tasks with proven optimal convergence and explicit bounds on performance gains.

Findings

Achieves optimal rate of convergence in cost-sensitive active learning.

Explicitly characterizes gain over passive learning based on boundary probability-mass.

Provides matching lower bounds up to logarithmic factors.

Abstract

Cost-sensitive learning is a common type of machine learning problem where different errors of prediction incur different costs. In this paper, we design a generic nonparametric active learning algorithm for cost-sensitive classification. Based on the construction of confidence bounds for the expected prediction cost functions of each label, our algorithm sequentially selects the most informative vector points. Then it interacts with them by only querying the costs of prediction that could be the smallest. We prove that our algorithm attains optimal rate of convergence in terms of the number of interactions with the feature vector space. Furthermore, in terms of a general version of Tsybakov's noise assumption, the gain over the corresponding passive learning is explicitly characterized by the probability-mass of the boundary decision. Additionally, we prove the near-optimality of obtained upper bounds by providing matching (up to logarithmic factor) lower bounds.