On Aggregation in Ensembles of Multilabel Classifiers

TL;DR

This paper introduces a formal framework for ensemble multilabel classification, distinguishing two main approaches—predict-then-combine and combine-then-predict—and demonstrates their advantages over standard voting methods through experiments.

Contribution

The paper formalizes ensemble multilabel classification with two approaches, enabling tailored optimization for different loss functions, and empirically shows their superiority over traditional voting techniques.

Findings

CTP performs well for decomposable loss functions.

PCT is better suited for non-decomposable losses.

Both approaches outperform standard voting methods.

Abstract

While a variety of ensemble methods for multilabel classification have been proposed in the literature, the question of how to aggregate the predictions of the individual members of the ensemble has received little attention so far. In this paper, we introduce a formal framework of ensemble multilabel classification, in which we distinguish two principal approaches: "predict then combine" (PTC), where the ensemble members first make loss minimizing predictions which are subsequently combined, and "combine then predict" (CTP), which first aggregates information such as marginal label probabilities from the individual ensemble members, and then derives a prediction from this aggregation. While both approaches generalize voting techniques commonly used for multilabel ensembles, they allow to explicitly take the target performance measure into account. Therefore, concrete instantiations of CTP and PTC can be tailored to concrete loss functions. Experimentally, we show that standard voting techniques are indeed outperformed by suitable instantiations of CTP and PTC, and provide some evidence that CTP performs well for decomposable loss functions, whereas PTC is the better choice for non-decomposable losses.