eCite Digital Repository

Probabilistic approaches for music similarity using restricted Boltzmann machines

Citation

Tran, SN and Ngo, S and Garcez, Ad, Probabilistic approaches for music similarity using restricted Boltzmann machines, Neural Computing and Applications, 32, (8) pp. 3999-4008. ISSN 0941-0643 (2020) [Refereed Article]

Copyright Statement

Copyright 2019 Springer-Verlag London Ltd, part of Springer Nature

DOI: doi:10.1007/s00521-019-04106-y

Abstract

In music informatics, there has been increasing attention to relative similarity as it plays a central role in music retrieval, recommendation, and musicology. Most approaches for relative similarity are based on distance metric learning, in which similarity relationship is modelled by a parameterised distance function. Normally, these parameters can be learned by solving a constrained optimisation problem using kernel-based methods. In this paper, we study the use of restricted Boltzmann machines (RBMs) in similarity modelling. We take advantage of RBM as a probabilistic neural network to assign a true hypothesis "x is more similar to y than to z" with a higher probability. Such model can be trained by maximising the true hypotheses while, at the same time, minimising the false hypotheses using a stochastic method. Alternatively, we show that learning similarity relations can be done deterministically by minimising the free energy function of a bipolar RBM or using a classification approach. In the experiments, we evaluate our proposed approaches on music scripts extracted from MagnaTagATune dataset. The results show that an energy-based optimisation approach with bipolar RBM can achieve better performance than other methods, including support vector machine and machine learning rank which are the state-of-the-art for this dataset.

Item Details

Item Type:Refereed Article
Keywords:music similarity, restricted Boltzmann machines, machine learning
Research Division:Information and Computing Sciences
Research Group:Artificial intelligence
Research Field:Intelligent robotics
Objective Division:Information and Communication Services
Objective Group:Information systems, technologies and services
Objective Field:Information systems, technologies and services not elsewhere classified
UTAS Author:Tran, SN (Dr Son Tran)
ID Code:139116
Year Published:2020
Web of Science® Times Cited:1
Deposited By:Information and Communication Technology
Deposited On:2020-05-27
Last Modified:2020-08-17
Downloads:0

Repository Staff Only: item control page