Hybrid Losses for Hierarchical Embedding Learning

TL;DR

This paper introduces hybrid loss functions for hierarchical embedding learning that improve classification, retrieval, and generalisation by leveraging label hierarchies and multi-task learning.

Contribution

It proposes novel hybrid loss functions and metrics for hierarchical embedding learning, enhancing model performance and generalisation to unseen classes.

Findings

Hybrid losses outperform previous methods in classification accuracy.

Enhanced embedding space structure and retrieval performance.

Improved generalisation to unseen classes in hierarchical datasets.

Abstract

In traditional supervised learning, the cross-entropy loss treats all incorrect predictions equally, ignoring the relevance or proximity of wrong labels to the correct answer. By leveraging a tree hierarchy for fine-grained labels, we investigate hybrid losses, such as generalised triplet and cross-entropy losses, to enforce similarity between labels within a multi-task learning framework. We propose metrics to evaluate the embedding space structure and assess the model's ability to generalise to unseen classes, that is, to infer similar classes for data belonging to unseen categories. Our experiments on OrchideaSOL, a four-level hierarchical instrument sound dataset with nearly 200 detailed categories, demonstrate that the proposed hybrid losses outperform previous works in classification, retrieval, embedding space structure, and generalisation.