Rizos, Georgios, Lawson, Jenna, Mitchell,, Simon L., Shah, Pranay, Wen, Xin, Banks-Leite, Cristina, Ewers, Robert, Schuller, Björn W (2024) Propagating variational model uncertainty for bioacoustic call label smoothing. Patterns (New York, N.Y.), 5 (3). Article Number 100932. E-ISSN 2666-3899. (doi:10.1016/j.patter.2024.100932) (KAR id:105474)
PDF
Publisher pdf
Language: English
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
|
|
Download this file (PDF/2MB) |
Preview |
Request a format suitable for use with assistive technology e.g. a screenreader | |
Official URL: https://doi.org/10.1016/j.patter.2024.100932 |
Abstract
Along with propagating the input toward making a prediction, Bayesian neural networks also propagate uncertainty. This has the potential to guide the training process by rejecting predictions of low confidence, and recent variational Bayesian methods can do so without Monte Carlo sampling of weights. Here, we apply sample-free methods for wildlife call detection on recordings made via passive acoustic monitoring equipment in the animals' natural habitats. We further propose uncertainty-aware label smoothing, where the smoothing probability is dependent on sample-free predictive uncertainty, in order to downweigh data samples that should contribute less to the loss value. We introduce a bioacoustic dataset recorded in Malaysian Borneo, containing overlapping calls from 30 species. On that dataset, our proposed method achieves an absolute percentage improvement of around 1.5 points on area under the receiver operating characteristic (AU-ROC), 13 points in F1, and 19.5 points in expected calibration error (ECE) compared to the point-estimate network baseline averaged across all target classes. [Abstract copyright: © 2024 The Authors.
Item Type: | Article |
---|---|
DOI/Identification number: | 10.1016/j.patter.2024.100932 |
Uncontrolled keywords: | variational Bayesian deep learning; uncertainty propagation; adaptive label smoothing; epistemic uncertainty; calibrated deep learning; bioacoustics; wildlife call detection; passive acoustic monitoring; machine audition |
Subjects: | H Social Sciences |
Divisions: | Divisions > Division of Human and Social Sciences > School of Anthropology and Conservation > DICE (Durrell Institute of Conservation and Ecology) |
Funders: | Engineering and Physical Sciences Research Council (https://ror.org/0439y7842) |
SWORD Depositor: | JISC Publications Router |
Depositing User: | JISC Publications Router |
Date Deposited: | 11 Apr 2024 14:13 UTC |
Last Modified: | 12 Apr 2024 07:56 UTC |
Resource URI: | https://kar.kent.ac.uk/id/eprint/105474 (The current URI for this page, for reference purposes) |
- Link to SensusAccess
- Export to:
- RefWorks
- EPrints3 XML
- BibTeX
- CSV
- Depositors only (login required):