Skip to Main content Skip to Navigation
Conference papers

Domain adaptation with optimal transport improves EEG sleep stage classifiers

Abstract : —Low sample size and the absence of labels on certain data limits the performances of predictive algorithms. To overcome this problem, it is sometimes possible to learn a model on a large labeled auxiliary dataset. Yet, this assumes that the two datasets exhibit similar statistical properties which is rarely the case in practice: there is a discrepancy between the large dataset, called the source, and the dataset of interest, called the target. Improving the prediction performance on the target domain by reducing the distribution discrepancy, between the source and the target domains, is known as Domain Adaptation (DA). Presently, Optimal transport DA (OTDA) methods yield state-of-the-art performances on several DA problems. In this paper, we consider the problem of sleep stage classification, and use OTDA to improve the performances of a convolutional neural network. We use features learnt from the electroencephalogram (EEG) and the electrooculogram (EOG) signals. Our results demonstrate that the method significantly improves the network predictions on the target data.
Complete list of metadatas

Cited literature [22 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01814190
Contributor : Stanislas Chambon <>
Submitted on : Wednesday, June 13, 2018 - 4:30:27 AM
Last modification on : Tuesday, May 5, 2020 - 11:50:13 AM
Document(s) archivé(s) le : Friday, September 14, 2018 - 3:13:45 PM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01814190, version 1

Citation

Stanislas Chambon, Mathieu N. Galtier, Alexandre Gramfort. Domain adaptation with optimal transport improves EEG sleep stage classifiers. Pattern Recognition in Neuroimaging, Jun 2018, Singapour, Singapore. ⟨hal-01814190⟩

Share

Metrics

Record views

408

Files downloads

875