Bayesian Inference for Correlated Human Experts and Classifiers

TL;DR

This paper introduces a Bayesian framework that efficiently combines correlated human expert opinions and pre-trained classifiers to reduce querying costs while maintaining high accuracy in classification tasks.

Contribution

It presents a novel Bayesian approach modeling expert correlation with a joint latent space, enabling efficient inference and query optimization in classification problems.

Findings

Significant reduction in human query costs in medical and image classification tasks

Maintained high accuracy with fewer expert queries compared to baselines

Effective modeling of expert correlation improves inference quality

Abstract

Applications of machine learning often involve making predictions based on both model outputs and the opinions of human experts. In this context, we investigate the problem of querying experts for class label predictions, using as few human queries as possible, and leveraging the class probability estimates of pre-trained classifiers. We develop a general Bayesian framework for this problem, modeling expert correlation via a joint latent representation, enabling simulation-based inference about the utility of additional expert queries, as well as inference of posterior distributions over unobserved expert labels. We apply our approach to two real-world medical classification problems, as well as to CIFAR-10H and ImageNet-16H, demonstrating substantial reductions relative to baselines in the cost of querying human experts while maintaining high prediction accuracy.