A Universal Unbiased Method for Classification from Aggregate Observations

TL;DR

This paper introduces a universal, unbiased classification method from aggregate data, enabling effective learning without individual labels, applicable across various loss functions and validated by extensive experiments.

Contribution

It presents a novel unbiased risk estimator for classification from aggregate observations, compatible with arbitrary loss functions, advancing the field of privacy-preserving and cost-effective learning.

Findings

Method achieves unbiased risk estimation for arbitrary losses.

Demonstrates superior performance across diverse CFAO tasks.

Ensures risk consistency and broad applicability.

Abstract

In conventional supervised classification, true labels are required for individual instances. However, it could be prohibitive to collect the true labels for individual instances, due to privacy concerns or unaffordable annotation costs. This motivates the study on classification from aggregate observations (CFAO), where the supervision is provided to groups of instances, instead of individual instances. CFAO is a generalized learning framework that contains various learning problems, such as multiple-instance learning and learning from label proportions. The goal of this paper is to present a novel universal method of CFAO, which holds an unbiased estimator of the classification risk for arbitrary losses -- previous research failed to achieve this goal. Practically, our method works by weighing the importance of each label for each instance in the group, which provides purified supervision for the classifier to learn. Theoretically, our proposed method not only guarantees the risk consistency due to the unbiased risk estimator but also can be compatible with arbitrary losses. Extensive experiments on various problems of CFAO demonstrate the superiority of our proposed method.