AEMLO: AutoEncoder-Guided Multi-Label Oversampling

TL;DR

AEMLO introduces an AutoEncoder-guided oversampling method that generates diverse synthetic samples for imbalanced multi-label datasets, improving classifier performance over existing techniques.

Contribution

The paper proposes a novel AutoEncoder-based oversampling approach specifically designed for multi-label imbalance, with a tailored objective function for better synthetic sample generation.

Findings

AEMLO outperforms state-of-the-art oversampling methods in empirical tests.

The method effectively generates diverse synthetic samples for imbalanced multi-label data.

AEMLO improves multi-label classifier performance on benchmark datasets.

Abstract

Class imbalance significantly impacts the performance of multi-label classifiers. Oversampling is one of the most popular approaches, as it augments instances associated with less frequent labels to balance the class distribution. Existing oversampling methods generate feature vectors of synthetic samples through replication or linear interpolation and assign labels through neighborhood information. Linear interpolation typically generates new samples between existing data points, which may result in insufficient diversity of synthesized samples and further lead to the overfitting issue. Deep learning-based methods, such as AutoEncoders, have been proposed to generate more diverse and complex synthetic samples, achieving excellent performance on imbalanced binary or multi-class datasets. In this study, we introduce AEMLO, an AutoEncoder-guided Oversampling technique specifically designed for tackling imbalanced multi-label data. AEMLO is built upon two fundamental components. The first is an encoder-decoder architecture that enables the model to encode input data into a low-dimensional feature space, learn its latent representations, and then reconstruct it back to its original dimension, thus applying to the generation of new data. The second is an objective function tailored to optimize the sampling task for multi-label scenarios. We show that AEMLO outperforms the existing state-of-the-art methods with extensive empirical studies.