Learning from Concealed Labels

TL;DR

This paper introduces a novel learning setting that protects sensitive labels by concealing them during data collection, enabling accurate multi-class classification without exposing sensitive information.

Contribution

It proposes an unbiased estimator for learning from concealed labels, ensuring accurate classification and privacy protection under mild assumptions.

Findings

The method achieves accurate classification on synthetic and real-world datasets.

The estimator bounds the error and attains optimal parametric convergence rate.

Experiments validate the effectiveness of concealed label learning in privacy-sensitive scenarios.

Abstract

Annotating data for sensitive labels (e.g., disease, smoking) poses a potential threats to individual privacy in many real-world scenarios. To cope with this problem, we propose a novel setting to protect privacy of each instance, namely learning from concealed labels for multi-class classification. Concealed labels prevent sensitive labels from appearing in the label set during the label collection stage, which specifies none and some random sampled insensitive labels as concealed labels set to annotate sensitive data. In this paper, an unbiased estimator can be established from concealed data under mild assumptions, and the learned multi-class classifier can not only classify the instance from insensitive labels accurately but also recognize the instance from the sensitive labels. Moreover, we bound the estimation error and show that the multi-class classifier achieves the optimal parametric convergence rate. Experiments demonstrate the significance and effectiveness of the proposed method for concealed labels in synthetic and real-world datasets.