Privileged Multi-label Learning

TL;DR

This paper introduces Privileged Multi-label Learning (PrML), a novel approach that leverages privileged label features and low-rank constraints to improve multi-label classification performance by exploiting label relationships.

Contribution

PrML combines privileged label features with low-rank constraints, providing a new framework that enhances multi-label learning by explicitly modeling label relationships and improving accuracy.

Findings

PrML significantly outperforms competing methods on benchmark datasets.

The algorithm is efficiently solvable using dual coordinate descent with iterative optimization.

Privileged label features lead to notable performance improvements in multi-label classification.

Abstract

This paper presents privileged multi-label learning (PrML) to explore and exploit the relationship between labels in multi-label learning problems. We suggest that for each individual label, it cannot only be implicitly connected with other labels via the low-rank constraint over label predictors, but also its performance on examples can receive the explicit comments from other labels together acting as an \emph{Oracle teacher}. We generate privileged label feature for each example and its individual label, and then integrate it into the framework of low-rank based multi-label learning. The proposed algorithm can therefore comprehensively explore and exploit label relationships by inheriting all the merits of privileged information and low-rank constraints. We show that PrML can be efficiently solved by dual coordinate descent algorithm using iterative optimization strategy with cheap updates. Experiments on benchmark datasets show that through privileged label features, the performance can be significantly improved and PrML is superior to several competing methods in most cases.