Scalable Multi-Output Label Prediction: From Classifier Chains to Classifier Trellises

TL;DR

This paper introduces the classifier trellis (CT) method, a scalable approach for multi-label classification that maintains competitive performance while efficiently handling datasets with thousands of labels.

Contribution

The paper proposes the classifier trellis (CT) method, improving scalability over traditional classifier chains for large multi-label datasets.

Findings

CT is highly competitive on standard multi-label problems.

CT scales efficiently to datasets with thousands of labels.

It overcomes the computational bottleneck of traditional chain methods.

Abstract

Multi-output inference tasks, such as multi-label classification, have become increasingly important in recent years. A popular method for multi-label classification is classifier chains, in which the predictions of individual classifiers are cascaded along a chain, thus taking into account inter-label dependencies and improving the overall performance. Several varieties of classifier chain methods have been introduced, and many of them perform very competitively across a wide range of benchmark datasets. However, scalability limitations become apparent on larger datasets when modeling a fully-cascaded chain. In particular, the methods' strategies for discovering and modeling a good chain structure constitutes a mayor computational bottleneck. In this paper, we present the classifier trellis (CT) method for scalable multi-label classification. We compare CT with several recently proposed classifier chain methods to show that it occupies an important niche: it is highly competitive on standard multi-label problems, yet it can also scale up to thousands or even tens of thousands of labels.