RLSEP: Learning Label Ranks for Multi-label Classification

TL;DR

This paper introduces RLSEP, a novel method for multi-label ranking that leverages true label orderings to improve ranking accuracy, outperforming existing methods on synthetic and real datasets.

Contribution

RLSEP proposes a new loss function that directly incorporates true label orderings, enhancing multi-label ranking performance beyond previous approaches.

Findings

Achieves state-of-the-art results on synthetic datasets.

Shows significant improvements on real-world ranked datasets.

Demonstrates generalizability across various multi-label tasks.

Abstract

Multi-label ranking maps instances to a ranked set of predicted labels from multiple possible classes. The ranking approach for multi-label learning problems received attention for its success in multi-label classification, with one of the well-known approaches being pairwise label ranking. However, most existing methods assume that only partial information about the preference relation is known, which is inferred from the partition of labels into a positive and negative set, then treat labels with equal importance. In this paper, we focus on the unique challenge of ranking when the order of the true label set is provided. We propose a novel dedicated loss function to optimize models by incorporating penalties for incorrectly ranked pairs, and make use of the ranking information present in the input. Our method achieves the best reported performance measures on both synthetic and real world ranked datasets and shows improvements on overall ranking of labels. Our experimental results demonstrate that our approach is generalizable to a variety of multi-label classification and ranking tasks, while revealing a calibration towards a certain ranking ordering.