Unsupervised Ensemble Learning with Dependent Classifiers

TL;DR

This paper develops unsupervised methods to detect dependencies among classifiers and improve ensemble learning accuracy when the common assumption of classifier independence is violated.

Contribution

It introduces a statistical model for dependent classifiers and novel methods for detecting dependencies and constructing better meta-learners in unsupervised ensemble learning.

Findings

Dependence detection improves ensemble accuracy.

The proposed methods outperform traditional independence-based approaches.

Results on real datasets demonstrate practical effectiveness.

Abstract

In unsupervised ensemble learning, one obtains predictions from multiple sources or classifiers, yet without knowing the reliability and expertise of each source, and with no labeled data to assess it. The task is to combine these possibly conflicting predictions into an accurate meta-learner. Most works to date assumed perfect diversity between the different sources, a property known as conditional independence. In realistic scenarios, however, this assumption is often violated, and ensemble learners based on it can be severely sub-optimal. The key challenges we address in this paper are:\ (i) how to detect, in an unsupervised manner, strong violations of conditional independence; and (ii) construct a suitable meta-learner. To this end we introduce a statistical model that allows for dependencies between classifiers. Our main contributions are the development of novel unsupervised methods to detect strongly dependent classifiers, better estimate their accuracies, and construct an improved meta-learner. Using both artificial and real datasets, we showcase the importance of taking classifier dependencies into account and the competitive performance of our approach.