Learning about individuals from group statistics

TL;DR

This paper introduces a probabilistic approach to infer individual labels from group-level statistics, advancing multiple-instance learning by estimating individual classifications from aggregate data.

Contribution

It presents a novel probabilistic model and an efficient MCMC training algorithm for learning individual labels from group statistics, improving upon existing methods.

Findings

Effective on synthetic data

Demonstrated on real-world object recognition data

Outperforms baseline methods

Abstract

We propose a new problem formulation which is similar to, but more informative than, the binary multiple-instance learning problem. In this setting, we are given groups of instances (described by feature vectors) along with estimates of the fraction of positively-labeled instances per group. The task is to learn an instance level classifier from this information. That is, we are trying to estimate the unknown binary labels of individuals from knowledge of group statistics. We propose a principled probabilistic model to solve this problem that accounts for uncertainty in the parameters and in the unknown individual labels. This model is trained with an efficient MCMC algorithm. Its performance is demonstrated on both synthetic and real-world data arising in general object recognition.