Improving Detection of Rare Nodes in Hierarchical Multi-Label Learning

TL;DR

This paper introduces a weighted loss function for hierarchical multi-label classification that improves detection of rare nodes, leading to significant gains in recall and F1 score, especially in challenging scenarios.

Contribution

It proposes a novel loss combining imbalance and focal weighting to better detect rare hierarchical nodes in neural networks.

Findings

Recall improved by up to five times on benchmarks.

Statistically significant F1 score gains.

Effective even with limited data or suboptimal encoders.

Abstract

In hierarchical multi-label classification, a persistent challenge is enabling model predictions to reach deeper levels of the hierarchy for more detailed or fine-grained classifications. This difficulty partly arises from the natural rarity of certain classes (or hierarchical nodes) and the hierarchical constraint that ensures child nodes are almost always less frequent than their parents. To address this, we propose a weighted loss objective for neural networks that combines node-wise imbalance weighting with focal weighting components, the latter leveraging modern quantification of ensemble uncertainties. By emphasizing rare nodes rather than rare observations (data points), and focusing on uncertain nodes for each model output distribution during training, we observe improvements in recall by up to a factor of five on benchmark datasets, along with statistically significant gains in $F_{1}$ score. We also show our approach aids convolutional networks on challenging tasks, as in situations with suboptimal encoders or limited data.