Boosting Generalization in Bio-Signal Classification by Learning the Phase-Amplitude Coupling

Lemkhenter, Abdelhak; Favaro, Paolo (2021). Boosting Generalization in Bio-Signal Classification by Learning the Phase-Amplitude Coupling. In: Akata, Zeynep; Geiger, Andreas; Sattler, Torsten (eds.) German Conference on Pattern Recognition DAGM-GCPR 2020. Lecture Notes in Computer Science: Vol. 12544 (pp. 72-85). Springer 10.1007/978-3-030-71278-5_6

[img] Text
2009.07664.pdf - Accepted Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (1MB)

Various hand-crafted features representations of bio-signals rely primarily on the amplitude or power of the signal in specific frequency bands. The phase component is often discarded as it is more sample specific, and thus more sensitive to noise, than the amplitude. However, in general, the phase component also carries information relevant to the underlying biological processes. In fact, in this paper we show the benefits of learning the coupling of both phase and amplitude components of a bio-signal. We do so by introducing a novel self-supervised learning task, which we call Phase-Swap, that detects if bio-signals have been obtained by merging the amplitude and phase from different sources. We show in our evaluation that neural networks trained on this task generalize better across subjects and recording sessions than their fully supervised counterpart.

Item Type:

Conference or Workshop Item (Paper)

Division/Institute:

08 Faculty of Science > Institute of Computer Science (INF)

UniBE Contributor:

Lemkhenter, Abdelhak, Favaro, Paolo

Subjects:

000 Computer science, knowledge & systems
500 Science > 510 Mathematics

ISBN:

978-3-030-71277-8

Series:

Lecture Notes in Computer Science

Publisher:

Springer

Language:

English

Submitter:

Abdelhak Lemkhenter

Date Deposited:

31 Mar 2021 16:05

Last Modified:

05 Dec 2022 15:48

Publisher DOI:

10.1007/978-3-030-71278-5_6

ArXiv ID:

2009.07664

BORIS DOI:

10.7892/boris.153029

URI:

https://boris.unibe.ch/id/eprint/153029

Actions (login required)

Edit item Edit item
Provide Feedback