Rank-Based Loss For Learning Hierarchical Representations

IEEE International Conference on Acoustics Speech and Signal Processing, pp. 3623–3627

Abstract

Hierarchical taxonomies are common in many contexts, and they are a very natural structure humans use to organise information. In machine learning, the family of methods that use this "extra" information is called hierarchical classification. However, applied to audio classification, this remains relatively unexplored. Here we focus on how to integrate the hierarchical information of a problem to learn embeddings representative of the hierarchical relationships. Previously, triplet loss has been proposed to address this problem, however it presents some issues like requiring the careful construction of the triplets, and being limited in the extent of hierarchical information it uses at each iteration. In this work we propose a rank based loss function that uses hierarchical information and translates this into a rank ordering of target distances between the examples. We show that rank based loss is suitable to learn hierarchical representations of the data. By testing on unseen fine level classes we show that this method is also capable of learning hierarchically correct representations of the new classes. Rank based loss has two promising aspects, it is generalisable to hierarchies with any number of levels, and is capable of dealing with data with incomplete hierarchical labels.

Authors 2

1. Inês Nolasco

   Queen Mary University of London

   Affiliation as printed

   Queen Mary University of London,Centre for Digital Music (C4DM),London,UK

   Centre for Digital Music (C4DM), Queen Mary University of London, London, UK

2. Dan Stowell Aachen Naturalis Biodiversity Center

   Tilburg University · Naturalis Biodiversity Center

   Affiliation as printed

   Tilburg University,Tilburg,The Netherlands

   Naturalis Biodiversity Centre, Leiden, The Netherlands

   Tilburg University, Tilburg, The Netherlands