University of Surrey

Test tubes in the lab Research in the ATI Dance Research

Learning with Out-of-Distribution Data for Audio Classification

Iqbal, Turab, Cao, Yin, Kong, Qiuqiang, Plumbley, Mark D. and Wang, Wenwu (2020) Learning with Out-of-Distribution Data for Audio Classification In: International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2020, May 4 to 8, 2020, Barcelona, Spain.

manuscript.pdf - Version of Record

Download (216kB) | Preview


In supervised machine learning, the assumption that training data is labelled correctly is not always satisfied. In this paper, we investigate an instance of labelling error for classification tasks in which the dataset is corrupted with out-of-distribution (OOD) instances: data that does not belong to any of the target classes, but is labelled as such. We show that detecting and relabelling certain OOD instances, rather than discarding them, can have a positive effect on learning. The proposed method uses an auxiliary classifier, trained on data that is known to be in-distribution, for detection and relabelling. The amount of data required for this is shown to be small. Experiments are carried out on the FSDnoisy18k audio dataset, where OOD instances are very prevalent. The proposed method is shown to improve the performance of convolutional neural networks by a significant margin. Comparisons with other noise-robust techniques are similarly encouraging.

Item Type: Conference or Workshop Item (Conference Paper)
Divisions : Faculty of Engineering and Physical Sciences > Electronic Engineering
Authors :
Plumbley, Mark
Date : 24 January 2020
Uncontrolled Keywords : Audio classification, out-of-distribution, convolutional neural network, pseudo-labelling
Depositing User : James Marshall
Date Deposited : 18 Feb 2020 11:38
Last Modified : 19 Feb 2020 15:37

Actions (login required)

View Item View Item


Downloads per month over past year

Information about this web site

© The University of Surrey, Guildford, Surrey, GU2 7XH, United Kingdom.
+44 (0)1483 300800