A Hidden Variables Approach to Multilabel Logistic Regression

TL;DR

This paper introduces a probabilistic multilabel classification model called MLRH that extends logistic regression by incorporating hidden variables, allowing for more flexible label modeling beyond one-hot encoding.

Contribution

The paper proposes a novel multilabel logistic regression model with hidden variables, enabling better modeling of label dependencies and relaxing traditional constraints.

Findings

Achieves competitive performance on benchmark datasets.

Extends logistic regression to multilabel problems with hidden variables.

Provides a probabilistic framework for multilabel classification.

Abstract

Multilabel classification is an important problem in a wide range of domains such as text categorization and music annotation. In this paper, we present a probabilistic model, Multilabel Logistic Regression with Hidden variables (MLRH), which extends the standard logistic regression by introducing hidden variables. Hidden variables make it possible to go beyond the conventional multiclass logistic regression by relaxing the one-hot-encoding constraint. We define a new joint distribution of labels and hidden variables which enables us to obtain one classifier for multilabel classification. Our experimental studies on a set of benchmark datasets demonstrate that the probabilistic model can achieve competitive performance compared with other multilabel learning algorithms.